host interaction for phage therapy against avian pathogenic

List of Abbreviations

Contribution to the understanding of the bacteriophage-

host interaction for phage therapy against avian pathogenic

Escherichia coli (APEC)

Patricia Espenhain Sørensen

This dissertation has been submitted in the fulfilment of the requirements for the degree of Doctor of

Philosophy (PhD) in Veterinary Sciences, Faculty of Veterinary Medicine, Ghent University and Ross

University School of Veterinary Medicine, 2022.

Promoters:

Prof. Dr. Patrick Butaye

Prof. Dr. An Garmyn

Prof. Dr. Hanne Ingmer

Faculty of Veterinary Medicine

Department of Pathobiology, Pharmacology and Zoological Medicine


Contribution to the understanding of the bacteriophage-host interaction for phage

therapy against avian pathogenic Escherichia coli (APEC)

PhD thesis, Ghent University and Ross University School of Veterinary Medicine, 2022

© Patricia Espenhain Sørensen

This research was funded by the European Union’s Horizon 2020 research and innovation program

under the Marie Skłodowska-Curie grant agreement no. 765147 and the Special Research Fund (BOF)

of Ghent University under the grant no. BOF.ITN.2021.0007.02. Conference attendance was supported

by Ghent University Mobility Fund. Additional funding was received from Augustinus Fonden and

Christian og Ottilia Brorsons Rejselegat.

Printed by: University Press, Belgium

Cover image: Electron microscopy pictures of Caudovirales coliphages by Liesbeth Couck.

Examination board

Prof. dr. Niek Sanders (Chair)

Prof. dr. Gunther Antonissen (Secretary)

Dr. Ilias Chantziaras

Dr. Steven Van Borm

Prof. dr. Rob Lavigne

Prof. dr. Felix Toka


“Impossible means that you haven’t found a solution yet”

Henry Ford (1863-1947)


i

Table of contents

List of Abbreviations ....................................................................................................... iii

Chapter 1: General Introduction...................................................................................... 1

1.1 Antibiotics and resistance ............................................................................................ 1

1.2 Bacteriophages ........................................................................................................... 4

1.2.1 General characteristics .......................................................................................... 4

1.2.2 Phage life cycle .................................................................................................... 6

1.2.3 Phage taxonomy ................................................................................................... 9

1.2.4 Phage diversity and signature genes.......................................................................11

1.2.5 Phage therapy......................................................................................................15

1.3 Phage-host interactions...............................................................................................21

1.3.1 Population growth dynamics.................................................................................22

1.3.2 Bacterial phage resistance.....................................................................................24

1.4 Avian pathogenic Escherichia coli (APEC) ..................................................................30

1.4.1 Diseases, transmission, and reservoirs ...................................................................30

1.4.2 Virulence factors .................................................................................................31

1.4.3 Strain typing and population genetics ....................................................................33

1.4.4 Current strategies to prevent and control APEC ......................................................33

1.4.5 Phage therapy against APEC infections .................................................................35

Chapter 2: Scientific Aims.............................................................................................. 61

Chapter 3: Experimental Studies ................................................................................... 63

3.1 New insights into the biodiversity of coliphages in the intestine of poultry ......................65

3.2 Classification of in vitro phage-host population growth dynamics................................. 115

3.3 Spontaneous phage resistance in avian pathogenic Escherichia coli .............................. 145

3.4 Schematic overview of the experimental studies and main findings .............................. 182

Chapter 4: General Discussion ..................................................................................... 185

4.1 Avian pathogenic E. coli (APEC) - The need for alternative treatment options............... 185

4.2 The remarkable diversity of E. coli-infecting phages ................................................... 186

4.2.1 Phage host spectrum .......................................................................................... 187

4.2.2 Hypothetical proteins of unknown function .......................................................... 188

4.3 Phage-host population growth dynamics .................................................................... 189

4.4 Phage resistance in APEC......................................................................................... 190

4.4.1 The cost of phage resistance ............................................................................... 193

ii

4.5 Acquisition and selection of suitable phages for phage therapy ..................................... 194

4.6 Conclusions and future perspectives .......................................................................... 196

Summary ...................................................................................................................... 207

Samenvatting ................................................................................................................ 211

Curriculum Vitae ......................................................................................................... 215

Bibliography ................................................................................................................. 217

Conference contributions ............................................................................................. 219

Acknowledgements ....................................................................................................... 221


iii


Abi Abortive infection

Acr Anti-CRISPR

AMR Antimicrobial resistance

AP Agar plate

APEC Avian pathogenic E. coli

AUC Area under the curve

BLAST Basic Local Alignment Search Tool

bp Base pairs

BREX Bacteriophage exclusion

BWA Burrows-Wheeler Aligner

C Cytosine

CARD Comprehensive Antibiotic Resistance Database

Cas CRISPR-associated proteins

CDS Coding sequence

CFU Colony forming unit

Coliphage E. coli-infecting phage

ColV Colicin V

CRISPR Clustered regularly interspaced short palindromic repeats

crRNA CRISPR RNA

crRNP crRNA-Cas protein

DLA Double-layer agar

DNA Deoxyribonucleic acid

ds Double-stranded

E. coli Escherichia coli

EPS Exopolysaccharides

EU European Union

ExPEC Extraintestinal E. coli

FAO Food and Agriculture Organization

G Guanine

GI Gastrointestinal

GMO Genetically modified organism


iv

GMP Good Manufacturing Practices

HGT Horizontal gene transfer

IC Intracranial

ICTV International Committee on Taxonomy of Viruses

IM Intramuscular

IT Intratracheal

Kbp Kilobase pairs

LB Lysogeny Broth

LCBs Local collinear blocks

LPS Lipopolysaccharide

MCP Major capsid protein

MDR Multidrug-resistant

MEGA Molecular evolutionary genetics analyses

MLST Multilocus sequence typing

MOI Multiplicity of infection

MTase Methyltransferase

MTP Major tail protein

NASP Northern Arizona SNP Pipeline

NCBI National Center for Biotechnology Information

NMDS Non-metric multidimensional scaling

OD Optical density

OIE World Organization for Animal Health

OMP Outer membrane protein

PCA Principal component analysis

PCR Polymerase chain reaction

PD Pharmacodynamics

Phage Bacteriophage

PHASTER PHAge Search Tool Enhanced Release

PK Pharmacokinetics

PFA Paraformaldehyde

PFU Plaque forming units

R-M Restriction-modification

RAST Rapid Annotation using Subsystem Technology

RBP Receptor-binding protein


v

REase Restriction endonuclease

RNA Ribonucleic acid

rpm Revolutions per minute

RT-qPCR Real-time quantitative PCR

SC Secondary culture

SEA-PHAGES Science Education Alliance-Phage Hunters Advancing Genomics and

Evolutionary Science

SNPs Single nucleotide polymorphisms

ST Sequence type

TA Toxin-antitoxin

TEM Transmission electron microscopy

TLS Terminase large subunit

TMP Tape measure protein

TPR Tetratricopeptide repeat

UPGMA Unweighted pair group method with arithmetic mean

WGS Whole-genome sequencing

WHO World Health Organization

WT Wild type


vi

Chapter 1: General Introduction

1


Chapter 1

General Introduction

1.1 Antibiotics and resistance

The modern era of antibiotics began with the discovery of penicillin by Sir Alexander Fleming

in 1928 [1, 2]. Since then, antibiotics have transformed modern medicine, veterinary as well as

human, and saved millions of lives. The “golden age” of antibiotics began in the 1940s and

continued for over four decades, with more than 20 classes of antibiotics being discovered and

introduced for clinical use [3–5]. During this period, bacterial resistance emerged, but was met

with minimal concern, as new compounds, often exhibiting better pharmacokinetics (PK) and

pharmacodynamics (PD), were quickly developed and provided alternative treatments [6, 7].

From the 1990s, as the number of novel antibiotics introduced steadily decreased, the

consequences of the link between antibiotic use/overuse and occurrence of resistance became

more apparent. This period has been described as a “dry pipeline” or “discovery void” in

antibiotic research and development, with fewer new drugs introduced, and with the majority

of these being either modified or combined versions of previously known compounds (Figure

1). As a result, many decades after the first patients were treated with antibiotics, more and

more bacterial infections are becoming a threat and difficult to treat once again [3, 8, 9].


2

Figure 1 | Timeline of the discovery of different antibiotic classes in clinical use. The “golden age” refers to the

period from 1940s to 1960 as one-half of the drugs commonly used today were discovered in this period. The

“discovery void” refers to the period from 1987 until today with only few new drugs introduced, and with majority

of these being semi- or fully synthetic [adapted from reference [8, 10, 11]].

The excessive use of antimicrobials in various areas, including clinical, industrial and

agricultural settings, and release into the environment for over half a century drives the

evolution of resistance and have generated a constant selective pressure for resistant bacterial

strains in all ecological niches [3, 12–14]. Studies have now demonstrated the direct association

between antibiotic consumption and the emergence and dissemination of resistant bacterial

strains [6, 7, 15–17]. The consecutive acquisition of antibiotic resistant traits has resulted in

the emergence of multidrug-resistant (MDR) and even pan-resistant pathogens. The resistance

can be caused by spontaneous mutations, recombination, or the acquisition of genes through

horizontal gene transfer (HGT), which is the major mechanisms involved in dissemination of

antibiotic resistance. In bacteria, HGT occurs mainly by conjugation, transformation and

transduction [18–20]. Conjugation is thought to be the main mode responsible for the spread

of antibiotic resistance [21]. During conjugation, deoxyribonucleic acid (DNA) is transferred

from the donor cell to a recipient through cell-to-cell contact and occurs either through plasmid

transfer or chromosomally integrated conjugation elements [19]. In transformation,

extracellular DNA is taken up by the bacterium from the environment and incorporated into

the genome, while in transduction, gene transfer is mediated by bacteriophages (phages) [18,

22, 23]. Also, resistance can automatically be transferred from one generation to the next


3

through replication (vertical gene transmission), unless the resistance-conferring element is lost

[24–26].

The global emergence of widespread antimicrobial resistance (AMR) has forced us to consider

the One Health approach to effectively control AMR and reduce the dissemination of resistance

genes between microorganisms [27]. The term “One Health” refers to the collaborative,

multisectoral, and transdisciplinary approach - working at the local, national, and global level

- to achieve optimal health outcomes by recognising the interdependency between people,

animals, plants, and their shared environment [28]. This holistic approach is required since

many of the antimicrobials used in human medicine are also used in veterinary medicine and

livestock production, as well as in plant production and their use drives selection of AMR,

regardless of the specific context in which they are used [27, 29]. Moreover, there is increasing

evidence that clinically relevant resistant bacteria and/or resistance genes are able to transfer

between animals and humans by overcoming both ecological and geographical barriers,

although the impact of this remains unclear [27, 29, 30]. Actions are needed to preserve the

continued effectiveness of existing antimicrobials by for example eliminating their

inappropriate use and by limiting the spread of infections through biosecurity measures [28,

31]. Remaining concerns are mass medication of animals in the animal production sector with

antimicrobials that are critically important for humans and the in-feed use of medically

important antimicrobials for growth promotion of healthy animals in some countries [28].

Numerous countries and several international agencies, such as the World Health Organization

(WHO), the World Organization for Animal Health (OIE) and the Food and Agriculture

Organization (FAO), have included a One Health approach within their action plans to assess,

control and prevent the spread of AMR as well as zoonotic diseases [28, 29]. Necessary actions

include improvement of antimicrobial use regulations, surveillance, infection control, animal

husbandry, and alternatives to antimicrobials [28]. Accordingly, as current antimicrobials

become increasingly inadequate, alternative treatment options are urgently needed. As such,

the use of phages as therapeutics (phage therapy) may help cope with the burden of

antimicrobial resistance [3, 12, 32–35].


4

1.2 Bacteriophages

1.2.1 General characteristics

Phages are viruses that specifically infect bacteria. They were discovered independently by

William Twort in 1915 and by Felix d’Herelle in 1917 who realised that they had antimicrobial

potential [36, 37]. d’Herelle used the term “bacteriophage” meaning “bacteria eater”, to

describe the organism’s antimicrobial ability. Phages are the most abundant organisms on

Earth, estimated about 4.8 x 1031 entities, and can be found in all known ecosystems, including

soil, wastewater, sewage water, seawater and in and on humans and animals [38–41]. Phages

outnumber their hosts by more than an order of magnitude, and are thought to play essential

roles in shaping the microbial communities, including driving the diversity, ecology and

evolution [38, 42–44]. Like other viruses, phages are unable to replicate independently of a

susceptible cellular host, and both their abundance and distribution are likely to be based on

that of their host. While some phages are able to infect hosts from different genera, families,

or orders, most studied phages are extremely specific and only capable of infecting a narrow

range of bacteria that are closely related [45–48]. The host range is determined by a

combination of various factors, including phage specificity, host attachment factors and

receptors, biochemical interactions during infection, presence of related (pro)phages in the

bacterial cell, and bacterial phage-resistance mechanisms [43, 45, 49–51]. In nature, phage host

range can be broadened or changed through mutation of receptor-binding proteins or exchange

and/or acquisition of new tail fiber genes by recombination, which may allow the phages to

move between related hosts [45, 52, 53].

The majority of phages, known to date, belong to the Caudovirales order (also known as the

tailed phages) (see section 1.2.3 Phage taxonomy) [54–56]. They have a double-stranded

(ds)DNA genome that range from about 18 to 500 Kilobase pairs (Kbp) [54]. The dsDNA is

enclosed in a polyhedral head, often being icosahedral, to which a tubular tailed is attached

(Figure 2). Most often, the head size of these phages range between about 45 nm and 170 nm

and the tail length between 3 nm and 825 nm. [54].


5

Figure 2 | Tailed phage. A) Negative staining transmission electron microscopy (TEM) images of tailed phage.

x60,000 magnification. Bar indicates 100 nm. Source: Lisbeth Couck, Department of Veterinary medical imaging

and small animal orthopaedics, Faculty of Veterinary Medicine, Ghent University. B) A schematic representation

of a tailed phage.

The tail morphology has traditionally been used for classification of the Caudovirales phages

into families: the Myoviridae family with a complex long contractile tail, the Siphoviridae

family with a long non-contractile tail, and the Podoviridae family with a short non-contractile

tail [56]. However, phage classification is currently undergoing extensive reorganisation,

primarily using genomic-based methods [57, 58]. Recently, it was suggested that these three

morphotypes are kept only as descriptors, and not as basis for establishment of phage families

as these would not be monophyletic [59].

The head-to-tail connecting region, termed connector or neck, ensures the interaction of the

phage capsid with its tail in all Caudovirales phages. It is often made of three different

components organised as consecutive rings: the portal protein and two head completion

proteins. The portal protein is located at the top of the “neck”/collar of the phage and is involved

in DNA packaging during assembly and release at the onset of infection [56, 60].

The composition of the Siphoviridae tail is rather simple and is based on three components: the

central TMP, the tail tube protein or major tail protein (MTP), and the tail terminator protein.

These components are also present and assembled in a similar way in Myoviridae tails, in

combination with the sheath protein that provides the contractile nature to the tail [61, 62]. At

the distal tail end, a special organelle (varying in size, composition, and morphology) dedicated

A) B)


6

to specific host recognition is found, and which controls phage specificity. The composition of

this organelle can be as simple as a tail tip or consists of a larger macromolecular complex

termed the baseplate. Despite these major conformational differences between tail tip and

baseplate, common scaffolding principles apply to both of these structural elements [60]. Also,

for many long-tail phages (Siphoviridae and Myoviridae phages), a similar consecutive open

reading frame order is often observed, including: the tail terminator, the MTP, the two tail

chaperones, the TMP, the baseplate hub, the tail-associated lysozyme or tail fiber, and varying

numbers of baseplate/tip proteins [60].

1.2.2 Phage life cycle

Phage adsorption to the bacterial host is one of the key aspects in phage life cycles [63]. When

a tailed phage encounters a susceptible host cell, the adsorption is facilitated by specific

recognition of host receptor surface proteins, such as lipopolysaccharide (LPS) (on the outer

membrane of Gram-negative bacteria), or other molecules (fimbria, flagella) on the bacterial

cell wall [64]. Successful recognition of bacterial surface receptor(s) leads to permanent phage

adhesion and allows for penetration of the bacterial cell wall using specialised enzymes,

followed by injection of the phage DNA thorough the cytoplasmic membrane and into the

cytoplasm. Depending on the phage type and host cell physiological condition, the phage will

enter a specific phage life cycle. There are four common phage life cycles, including lytic,

lysogenic, pseudolysogenic and chronic infections [42, 43, 63, 65]. As chronic infection is

typical for only filamentous phages (out of scope of this dissertation), this life cycle will not be

described in more detail but we refer the reader to the papers of [66, 67].

Virulent or obligate lytic phages are strict pathogens of the bacterial host. These phages can

only replicate through the lytic cycle and infection results in the production of new phage

particles and lysis of the host [42]. The lytic cycle includes infection, transcription, phage

replication, and particle assembly and release (Figure 3). After successful infection the phage

DNA is transcribed. The phage genome encodes early proteins, including endonucleases and

exonucleases to degrade host DNA. The phage takes over host metabolism to replicate,

transcribe and translate phage structural component-encoding genes. Phages have evolved a

variety of transcriptional control strategies that range from full dependence on host

transcription machinery to near-complete independence [65]. Moreover, phages can conserve

energy for infection by shutting off “non-essential” host process, such as host replication and


7

cell division [68, 69]. The efficiency of host hijacking differs between different phages.

Specialist phages infecting their preferred host seem more efficient compared to generalist

phages that infect multiple hosts. These generalist phages tend to have less efficient infections

and fail to completely suppress host translation and transcription [68]. Once all the structural

components have been translated, they are assembled into new phage particles, and the phage

DNA (or ribonucleic acid (RNA)) is packed into the capsid. Finally, the newly produced phages

are released from the bacterial host to the environment through lysis. The bacterial membrane

and cell wall disruption is facilitated by a combination of specific phage-encoded lysins, such

as holins and endolysins. Holins are small proteins that cause non-specific lesions in the

bacterial plasma membrane, allowing the endolysins to reach the peptidoglycan and attack the

murein layer of the bacterial cell wall. Phages infecting Gram-negative hosts can utilise

additional proteins, called spanins, that aid the lysis through destabilisation of the outer

membrane [70].

Temperate phages have the ability to switch between the lytic and lysogenic life cycle. Whether

the phage will follow the lytic or lysogenic pathway is decided at the start of each infection in

response to phage-, host-, and environmental factors [71]. Among others, host abundance,

defined by multiplicity of infection (MOI), may determine which pathway is followed; low

MOI favours lytic replication, whereas high MOI favours lysogeny [72–74]. Factors such as

host cell activity may play a role in phage replication, as conditions that cause reduced activity,

such as low nutrients or reduced host fitness favour lysogeny [71, 75]. Also, some phages use

a phage-encoded signal peptide to coordinate lysis-lysogeny initiation [76]. During the

lysogenic cycle the phage DNA integrates into the bacterial genome (or stays as a plasmid

inside the host cell), rather than replicates and produces new phages. Following successful

infection, the phage DNA is integrated into the bacterial genome as a prophage by a specific

phage-encoded DNA insertion enzyme called integrase. This integration includes breakage and

re-joining of the phage and bacterial host DNA. Following prophage integration, the bacterial

cell remains alive and continues to grow and replicate together with the prophage. The

prophage genes are replicated as part of the bacterial genome and are transmitted to the

daughter cells, resulting in a large population of bacteria infected with prophages [43, 77]. The

prophage genome is maintained by phage-encoded repressors, which controls expression of

genes required for prophage excision, but can be excised from the genome and enter the lytic

cycle when induced. The induction signals vary among phages but prophages are commonly

induced when the bacterial SOS response is activated due to exposure to stress or adverse


8

environmental conditions, leading to inactivation of repressors responsible for prophage

maintenance [78]. Stressors include changes in temperature, pH or nutrients, and exposure to

antibiotics, foreign DNA or DNA damaging agents (such as ultraviolet light) [71, 79].

Temperate phages may become lytic/virulent if their integrase gene is deleted or damaged by

mutation or genetic engineering. Alternatively, some prophages can influence the induction of

other prophages [71]. A small fraction of prophages in a population might spontaneously excise

from the chromosome and enter the lytic state without any apparent external triggers [80, 81].

The lysogenic cycle can be stable for thousands of bacterial generations and the phage may

alter the phenotype of the bacterium by expressing prophage-encoded genes (lysogenic

conversion). This can increase the fitness of the host, including increased pathogenicity, and

thus, also the survival rate of the phage [42, 82, 83]. Through lysogeny, phage genes are

maintained in bacterial hosts throughout microbial communities and more than 80% of

prokaryotic genomes are predicted to contain at least one prophage [38].

Figure 3 | Example of temperate coliphage life cycles. The phage attaches to a host cell and injects its DNA. Next,

certain factors, such as repressor proteins or antibiotics, determine whether the phage enter the lytic or the

lysogenic cycle. In the lytic cycle, transcription, translation, and replication of phage DNA are initiated using

bacterial materials and phage enzymes. Synthesised phage DNA and proteins are assembled into phage particles.

Finally, the bacterial cell lyses and the phages are released. In the lysogenic cycle, phage induction is repressed,

and the phage DNA integrates into the bacterial chromosome and as a prophage. Afterwards, the bacterium

replicates normally, copies the prophage and transmits it to daughter cells. However, if induced , the prophages

may excise from the bacterial chromosome and initiate the lytic cycle [modified from [77]].


9

Pseudolysogeny can be defined as the stage of stalled development of phage in a host cell [84].

It is often caused by unfavourable growth conditions for the bacterial host (such as starvation)

where there is insufficient energy available for the phage to initiate genetic expression and

replication [63]. After entering the host cell, the phage DNA resides inactive within the cell,

and the replication cycle is halted until environmental conditions improve [84].

Pseudolysogeny occurs in both lytic and temperate phages. While the lytic replication cycle is

simply stopped, the lysogenic infection may lead to two subpopulations of bacteria: lysogens

and phage-carrying cells, resulting in infected and non-infected cell lineages [63]. The

pseudolysogenic state may explain the long-term survival of phages in unfavourable

environments in nature [84, 85].

1.2.3 Phage taxonomy

In contrast to bacteria, no single conserved gene is present in all phages. As a consequence, the

taxonomic classification of phages is based on host range, physical characteristics, including

size, structure, and morphology, genetic makeup, and overall genomic similarity [86, 87].

Moreover, defining characteristics can be determined for each phage genus, including average

genome length and number of coding sequences (CDSs), percentage of shared CDSs, and the

presence of specific signature genes in genus member phages [88]. The phage classification

scheme is regularly updated, refined and approved by the International Committee on the

Taxonomy of Viruses (ICTV) [89]. In recent years several genome-based phage taxonomy

schemes have been introduced [87, 90] and taxonomy has changed considerably [57, 91][59].

Currently (October 29, 2021), more than 11.000 complete phage genomes have been included

in the National Center for Biotechnology information (NCBI) Nucleotide database. However,

despite a continuously increasing number of sequenced genomes, most phages remain

unclassified and poorly characterised.

Most E. coli-infecting phages, or named also coliphages, belong to the highly heterogeneous

Caudovirales order, which constitute ~94% of all known isolated phages [54, 58]. To this date,

this order contains 14 families of tailed phages with dsDNA genomes (Figure 4):

Ackermannviridae, Autographiviridae, Chaseviridae, Demerecviridae, Drexlerviridae,

Guelinviridae, Herelleviridae, Myoviridae, Podoviridae, Rountreeviridae, Salasmaviridae,

Schitoviridae, Siphoviridae and Zobellviridae [92]. Following the ICTV taxonomy

(https://talk.ictvonline.org/taxonomy/), these families comprise 73 subfamilies, 927 genera,

https://talk.ictvonline.org/taxonomy/


10

and 2814 species. In addition, this order includes a single genus with no designated family. The

number of coliphages in the Caudovirales order constitute ~13% (n=374) of the registered

species. Coliphage species are found in nine of the 14 families (Figure 4).

Figure 4 | The Caudovirales order according to ICTV taxonomy. Numbers in the Subfamily column indicate

number of subfamilies within each family. Numbers in the Genus column indicate number of genera within each

family. Numbers in the Species column indicate number of species within each family. Numbers in brackets

indicate the number of coliphage species. Figure based on numbers from ICTV taxonomy, accessed on October

28, 2021.

For tailed phages, it has been reported that conserved genes such as the terminase large subunit

(TLS), the portal protein and major capsid protein (MCP), can be used as phylogenetic markers

for the diversity as well as their evolutionary relationship [48, 93]. Furthermore, automated

classification of tailed phages can be done according to their neck organisation [94].


11

1.2.4 Phage diversity and signature genes

Characterisation of phage abundance and diversity traditionally involves phage culture-based

techniques and plaque assays [51, 95–97]. The advantage of these methods includes the viable

counts of phage particles and the potential for phenotypic characterisation of phages, including

host range determination. Challenges associated with these methods, include the requirement

of phages to produce plaques, that plaques can be formed under the plaquing conditions

employed, and the demand of a suitable host bacterium. Consequently, the re are strong

selective biases in determinations of phage environmental diversity using these traditional

methods [98]. Biases may be even stronger when methods of pre-enrichment and propagation

prior to plaquing are included. Also, there exists no guarantee that the host strain used is

susceptible to a fair representation of the phages in the environment of interest. Most phages

have been isolated using a single bacterial host strain [40, 99]. While this procedure is widely

used, it may likely produce narrow rather than broad host range phage. One way to obtain more

broad host range phages is using a sequential multiple host strains-approach during isolation

[47]. For isolation studies, suitable host strains can either be isolated specifically from the

environment of interest, or be a “model” host, such as a well-characterised laboratory strain,

chosen based on specific desirable traits [40]. Evidently, only phages that can infect the specific

host strain(s) will be identified, and accordingly, it is difficult to establish what proportion of

phages present in the environment are being isolated [42, 43]. For phage therapy (see section

1.2.5), testing on clinical bacterial isolates may be more relevant than testing on laboratory

strains [49].

Transmission electron microscopy (TEM) has traditionally been used for direct assessment of

the phenotypic diversity and abundance of phages. It offers powerful magnification and

provides information of surface features, shape, size, and structure. The number of detected

phage morphological types varies significantly between studies and might reflect the diversity

of different microbial communities. However, variations between studies might also reflect

low sensitivity as well as low specificity [40]. Tailed phage morphologies are unique and

different from other viruses. However, tailed phages with highly similar morphology may in

fact have very different genomes. Thus, nucleotide sequence information (preferably whole

genome sequence) is required to fully understand the diversity, relationships, and dynamics

among the members of any set of phages being compared [48, 56].

Recent advances in viral omics and high-throughput sequencing methods have enabled the

rapid discovery of various phages in numerous environments and have broadened our view of


12

phage abundance and diversity [56, 100]. Although these advances have expanded our

understanding of phage genomic diversity, they also revealed that we have only scratched the

surface of the abundance of phage diversity. It is predicted that more than 99% of viral genetic

diversity remains to be revealed [56, 101].

Whole-genome sequencing (WGS) is a comprehensive method for determination of the DNA

sequence of an organism’s genome. Decreasing sequencing cost and the ability to produce large

volumes of sequence data in a short amount of time make WGS a powerful tool for genomic

research. Different sequencing technologies are available. Short-read sequencing (such as the

Illumina platform), also referred to as second generation sequencing, offers the potential to

rapidly sequence hundreds of phage genomes with high accuracy (~99%). This sequencing

method generates high read counts of short reads (150-300 bp) within a single run producing

high coverage, and the base-by-base sequencing protocol enables the accurate data acquisition

[102, 103]. However, all short-read sequencing technologies have a common limitation – the

inability to assemble long stretches of DNA resulting in relatively fragmented genome contigs.

Long-read sequencing (such as Nanopore, and PacBio singe-molecule real-time sequencing),

also referred to as third generation sequencing, address the shortcomings of short-read

sequencing with read length of >10.000 on average [104]. Longer reads are especially useful

when sequencing complex genomic regions such as repeats and phages. However, these longer

reads are more prone to errors resulting in sequencing accuracy of ~92-97%. Though more

expensive, combining short-read and long-read sequencing has emerged as a promising

approach to overcome pitfalls associated with singe-technology use and generate fully resolved

and accurate genome assemblies [103, 105]. Different sequencing approaches can be applied

depending on the sequencing goal. Metagenomics is the sequencing of all DNA present in a

sample as opposed to sequencing just a single microorganism [106]. This approach has

extraordinary potential to improve our understanding of for example (complex) microbial

populations in their natural environments or primary sample, regardless of whether they belong

to microorganisms that can be cultured in the laboratory. Also, this isolation-free, culture-

independent method does not rely on amplification of specific genomic sequences, which can

otherwise introduce bias [107]. However, this approach does not provide high-resolution

needed for in-depth characterisation of single genomes and may produce biases towards certain

sequences rather than abundance [108].

Yet, despite the remarkable diversity of phages at the nucleotide sequence level and the lack of

a universal conserved marker gene found throughout all phage families, the structural proteins


13

that form viral particles show strong similarity and conservation, representing important

taxonomic characteristics [56, 109]. Accordingly, diversity and abundance can be assessed

using family-specific signature “marker” genes shared by all members [110, 111]. One of the

most conserved marker gene type of the tailed phages are the terminases. These genes are

phage-coded proteins that bind to and cut DNA. They consist of a large and a small subunit

with molecular weights of 44-73 and 10-45 kDa, respectively. The small subunit is responsible

for DNA recognition and binding, initiating the packaging of the viral genome. The larger

subunit (TLS) ensures DNA cutting, binding of the terminase to the connector, and DNA

translocation into the empty phage head (capsid) to finalise the packaging process [112, 113].

Moreover, the two conserved structural proteins: portal protein and MCP, involved in the phage

head-assembly process, appear to be universally present in the tailed phage genome. The use

of these “conserved” genes has been shown useful for the characterisation of Myoviridae

phages, including when used as signature marker genes either alone or in combination. The

T4-like g20 portal protein gene has been shown to be conserved among phages that inhabit

marine and freshwater environments as well as eukaryotic hosts [60, 110], and the T4-like gp23

MCP has been shown useful in assessing Myoviridae diversity [111, 114]. Also, multi-marker

gene studies have been done to assess phage diversity using a multilocus sequence typing

(MLST) scheme approach [111]. By using (degenerate) polymerase chain reaction (PCR)

primers specific for identified marker genes and sequencing of the amplified environmental

sequences, these marker genes offer new ways of providing estimates of diversity and

quantifying phage abundance without the cultivation-based complications. Based on those

marker gene studies, the global phage diversity far exceeds that represented by cultured isolates

[115–117]. However, despite providing insights into the overall distribution of specific phage

genes, even in the best cases, signature genes fail to capture the full diversity present in natural

communities [38]. While amplicon methods such as PCR are fast and low-cost, they can suffer

from amplification bias [118], which might be more apparent when examining less abundant

genes or genomes [119]. Marker gene abundance only reflects phage abundance if the gene is

present and detected. Scaffolding proteins and procapsid proteases are often but not universally

encoded. Head-tail joining or connecting proteins are likely present in all tailed phage virions

but are too diverse to be recognisable in all of them at present. As such, no universal primer

can be designed, making it cumbersome to design primers that target the whole phage diversity

[97]. Recent studies have shown that the different tail tape measure proteins (TMPs) correlates

well with clusters for Siphoviridae and Myoviridae. However, the short tails of the Podoviridae

have no TMP. Moreover, there is no known tail protein that is common to all types of tails, and


14

both phage DNA replication/metabolism and lysis mechanism are too diverse so that no

homologue proteins can be used universally for all phage families [48].

The diversification of phage genomes is driven by multiple mechanisms, including the

accumulation of mutations, gene acquisition and loss , and recombination events [48, 56, 120].

For tailed phages, especially HGT contributes to evolution as well as diversification of the

genomes [48]. The horizontal transfer of genetic material occurs via both homologues and non-

homologues recombination events, and both within and between phages as well as bacterial

hosts. Especially homologues recombination between genomes of co-infecting phages is

thought to be the main mechanism of HGT [121]. Temperate phages have been shown to

acquire DNA from defective prophages through homologues recombination [122]. Some

phages are able to exchange up to 79% of their genome [123, 124]. Phage genomes exhibit

genetic mosaicism with conserved genetic modules that encode exchangeable functional units

such as the virion coat-encoding genes, the virion tail-encoding genes, or the genome

integration proteins [123, 125]. The creation of this mosaicism is an ongoing process, driven

by the HGT and recombination events [52, 120, 122, 126, 127]. Such horizontal exchange

among phages complicates whole genome comparisons and might make any strictly

hierarchical classification scheme insufficient and potentially misleading if the exchange is

great enough [125]. Accordingly, single (marker) gene locus use, such as the terminase, is

usually not used independently, but rather in combination with other marker genes and/or other

approaches to minimise the potential impact of HGT events on the taxonomy [59]. However,

horizontal exchange does not appear to be rapid enough to destroy the overall relationships

within or between ‘phage types’ [48, 52, 99, 127]. Also, the rate of HGT is associated with

phage lifestyle: most phages with a high rate of HGT are temperate whereas the majority of

phages with low rates of HGT are strictly virulent [120, 128].

In silico analyses of phage genomes have shown that there is a large number of phage genes

whose function cannot be predicted due to very low or no similarity to already known genes

[129]. However, due to high degree of conservation of function-associated gene orders in

regions encoding morphogenesis modules in tailed phages, it is possible to identify protein

functions in the absence of detectable sequence similarity [60].


15

1.2.5 Phage therapy

Phage therapy is defined as the application of phages to treat or prevent bacterial infections

[130]. The therapeutic potential of phages was first discovered over a century ago, but the

discovery and widespread use of antibiotics led to a loss of interest in the therapeutic

application of phages [3]. Still, phage therapy research and application did continue in some

countries, as in Georgia (part of the former Soviet Union) and Poland, where phages were, and

continue to be, routinely used to treat a large number of diseases [3, 131]. Nowadays, we are

facing a worldwide increase in the prevalence of antibiotic resistant bacteria, and lack of

discovery of new antimicrobials, urging for alternative treatment options. This, along with

advances in modern molecular biology, biotechnology, and genetic engineering, have led to a

renewed interest in phage therapy [9, 132–134]. Furthermore, advances in bioinformatics and

WGS technologies, including genome sequencing of phages and entire microbiomes, have

broadened our understanding of phage taxonomy, host-specificity, population structure and

genomic evolution [38, 59, 129, 135–137]. Also, with recent advances the phage application

spectrum has been expanded to various medical, biotechnological and agricultural fields,

including the use of phages in phage therapy in humans and animals, surface dis infections,

bacterial detection, gene delivery, food bio-preservation and safety, biocontrol of food and

plant pathogens, and biofilm control [138–142].

Phages, as therapeutic agents, have numerous advantages that make them good alternatives or

supplements to antibiotics (Table 1). 1) New phages are often relatively easily discovered and

isolated due to the great biodiversity of phages in nature [143, 144]. Any environment that

contains the pathogen of interest is likely to contain phages that are able to infect and kill that

organism [51]. 2) Strictly lytic phages are by nature bactericidal [145]. Phages’ ability to

effectively eliminate bacterial pathogens in animals and humans has convincingly been

demonstrated, and doses as low as 102 plaque forming units (PFU) have been shown to be able

to prevent disease in animals by pathogenic E. coli [146, 147]. 3) As phages infect and kill

bacteria using mechanisms that differ from those of antibiotics, phages can be used to target

bacterial states such as biofilms, persistence, and bacteria that are antibiotic resistant [148].

The high number of bacteria present within the biofilm(s) facilitate rapid and efficient phage

infection of the host and consequent phage replication. Also, phages can produce specific

enzymes that degrade the extracellular matrix of the biofilm. Phages are able to infect persister

cells where they remain dormant, but re-activate when the host cells become metabolically

active [149, 150]. 4) Phages hijack multiple essential cellular processes, including DNA


16

replication, transcription, and translation upon infection [144]. 5) Due to the host specificity of

phages, they tend to only minimally disrupt the normal microflora by selectively targeting only

pathogens [151]. By contrast, many antibiotics, which tend to have broader spectrums of

activity, might cause damage to all bacterial cells independently of whether they are pathogenic

or commensal [152]. Such disturbance of normal microbiota can amongst others result in

diarrhoea as well as increased risk of secondary infections [40, 144]. Also, the relatively narrow

host range exhibited by most phages limits the risk of cross-resistance between different phages

[144, 153]. 6) As phages co-evolve with bacteria over time, the administered phage population

may evolve to (re-)infect the phage-resistant bacteria (an arms race), which is not possible for

antibiotics [132, 154, 155]. 7) If phage resistance should develop, careful choice of phage(s)

that select for resistant bacterial mutant types with lower fitness, such as reduced cell division

rates and pathogenicity traits expression, could be an advantage despite resistance development

[144]. Bacteria resistant to LPS-targeting phages, are typically reduced in both fitness and

virulence [132, 156]. 8) Phages are self-regulating and can increase in number over the course

of treatment, specifically at the site of infection, where most target bacteria are present. This

also allows for less frequent dosing and low-dosage use of phage compared to antibiotics [144].

9) Finally, phages are self-limiting as they will decrease in abundance as soon as susceptible

host cells are eliminated [152].

Despite the numerous advantages of phage therapy, phages still have limitations (Table 1).

Both regulatory and technical hurdles must be overcome before phage therapy can be fully

accepted in modern clinical practice. Not all phages make for good therapeutics. Good

therapeutic phage candidates should have a high potential to reach and then kill target bacteria

without negatively modify the environments to which they are applied. Such high “virulence”

includes good adsorption properties, high potential to evade bacterial defences, good

replication characteristics, and/or high fecundity (short latent period and large burst size) [157,

158]. These characteristics can be reasonably assured using phages that are obligately lytic,

viable, stable under typical storage conditions and temperatures, subject to appropriate efficacy

and safety studies, and, ideally, fully sequenced to, among others, confirm the absen ce of

bacterial virulence factors [132, 144, 159]. To achieve therapeutic efficacy, the phages applied

should be able to replicate (or, at least, infect) at the expense of their bacterial target faster than

they are removed from the site of treatment such as by the host immune system or by

environmental turnover (in vivo persistence) [157, 160].

Chapter 1: Genera l Introduction

17

Table 1 | Advantages and disadvantages of phage therapy

Trait Advantages Disadvantages

Bactericidal

agents

Lytic phages cause host cell lysis.

Active against Gram-positive and

Gram-negative bacteria, including

MDR-variants

Not accessible to intracellular pathogens

Specificity

Highly specific, minimal or no

disruption to normal microbial

community

Narrow host spectrum, host bacterium

needs to be identified.

Resistance

Phage-resistant mutants are often

less virulent, as phage receptors are

commonly associated with

pathogenicity. Able to evolve to

overcome bacterial resistance.

No cross-resistance to antibiotics

Risk of phage resistance development in

bacteria

Dosage

Simplifying dosage. Self-regulating

in proportion with target bacteria,

replicates at the site of infection

Depend on susceptible host for

replication. When target organism is not

present the phages will not replicate.

Perceived by the immune system as

invaders and can be rapidly degraded

Toxicity

Generally considered as safe due to

nucleic acid and protein

composition

Rapid lysis of bacteria may lead to the

release of endotoxins and induce

inflammatory immune response

Discovery Rapid and relatively easy discovered

due to their ubiquitous nature

Depend on susceptible host bacterium

for isolation and replication

Phage cocktails Can broaden host range and reduce

risk of resistance development

Lack of standardised guidelines to

generate phage cocktails

Implementation New regulatory framework for

phage therapy in Belgium.

Regulatory hurdles. Not accepted as

pharmaceutical drugs in most countries.

Difficulties in patenting

Bioengineering Can be genetically modified Requires host bacterium, expertise, and

technology

Other Exhibit anti-biofilm properties and

can target persister cells -

Information adapted from [3, 32, 144, 157, 161].


18

Temperate phages are usually avoided as their genomes may contain genes which alter the

phenotype of the host after infection (lysogenic conversion). These phages are capable of

generalised transduction, and thus, are able to transfer large amounts of bacterial DNA from

one host to another, including virulence and antibiotic resistance genes [130, 160]. Also,

administration of temperate phages may not result in an immediate bactericidal effect on the

target pathogen if the phages integrate as prophages. Furthermore, when integrated in the

bacterial chromosome, prophages may display superinfection immunity, making bacteria

resistant to further phage infections [162, 163].

Advances in the field of biotechnology, synthetic biology and genetic engineering facilitate

engineering of phages in various ways that potentially improve the antimicrobial properties of

the phages as well as create new strategies for fighting bacterial infections [32, 130, 163, 164].

Genetic engineering potentially improves phage efficiency. Phages engineered to express

biofilm matrix-degrading enzymes penetrates biofilms more readily than the non-engineered

wild type (WT) phage [165]. Phage host range can be altered to serve practical purposes.

Synthetic phage variants with altered host range have been constructed by swapping tail

component-encoding genes [166]. To bypass concerns regarding uncontrolled self -

amplification and sudden release of bacterial endotoxins upon lysis, phages can be engineered

to be non-replicative and used to deliver genes, interfering with important bacterial intracellular

processes, into specific bacterial populations through transduction [167]. As mentioned above,

strictly lytic phages might be preferable for phage therapy. However, some bacterial species

appear to produce only temperate phages [168]. Genetic engineering can be used to obtain

strictly lytic derivatives of temperate phages in which the repressor and/or integrases are

deleted. Also, gene encoding bacterial virulence factors and integrases can similarly be

removed [163]. However, considerable considerations have been given to the use, design, and

associated risks [169]. Also, there are still a lot of difficulties in engineering phages. Many

strategies require the ability to genetically modify the bacterial host(s) or to efficiently deliver

exogenous DNA into these hosts, which is still a challenge for many bacterial species [170].

Most phage genomes are too large (>20 Kbp) for easy in vitro manipulation and are lethal to

their bacterial host [167]. Temperate phages can be stably maintained in the bacterial genome,

enabling modification of the phage genomes using the same methodologies as those used for

engineering of bacterial genomes. However, genomes of virulent phages cannot be cloned

whole into bacteria for subsequent genetic modification. As a result, virulent phage genomes

are commonly edited using homologues recombination (allelic exchange), whereby the gene(s)


19

to be modified are cloned into a plasmid and modified as needed before being introduced to

the host bacterium. Homologues recombination efficiency might be low and many phage loci

cannot be cloned due to their adverse effect on bacteria [167]. Consequently, there is a strong

pressure for developing new phage genome engineering methods. Intracellular bacterial

pathogens constitute another limitation of phage therapy as phages do not have a mechanism

of entry into eukaryotic cells. Also, as a phage population can undergo rapid exponential

growth, widespread lysis of target bacteria may potentially release endotoxins that could be

harmful to the patient [132, 171]. Limited knowledge on the phage interaction with patient

immune system is another matter of concern [132, 172, 173]. The great specificity of phages

might represent a challenge to phage therapy as it requires the potential need of characterisation

of bacterial susceptibility prior to the phage application. Thus, it is essential to know exactly

which bacterial species is the causative agent of the infection, and success of phage therapy is

associated with careful choice of phage capable of infecting the causative bacterial agent [132].

However, genetic engineering can potentially address the phage specificity shortcomings and

increase the therapeutic potential of natural phages [167]. The second method of overcoming

a too narrow spectrum of activity is combining phages (so-called phage “cocktails”) [133, 152,

157, 160, 164, 174]. Emergence of phage resistant bacteria constitute another limitation to

phage therapy [175, 176], as bacteria are readily capable of evolving resistance to phages

through a variety of different mechanisms (see section 1.3.2).

Using a cocktail of phages might reduce the probability of phage-resistant bacteria emerging

as different types of phages infecting the same species and/or strains are present [177, 178].

Nevertheless, even resistance to phage cocktails have been described and as such cannot be

regarded as the most optimal solution [179, 180]. When designing phage cocktails the

following should be taken into account to overcome the emergence of resistance. First, the

mechanism(s) whereby phage resistance can evolve should be taken into consideration. By

knowing the specific mechanisms, phages that select for resistance associated with reduced

fitness and pathogenicity can be chosen. Also, phages with desired properties to overcome

these resistance mechanisms can be chosen. Second, the potential of bacteria to develop cross-

resistance to multiple phages should be considered. The risk of cross-resistance could be

overcome by combining phages utilising different receptors, as the fitness cost associated with

resistance development to all might be too high for the host bacterium and therefore unlikely.

Finally, it should be confirmed that the cocktail phages do not compete with one another and

as such reduce the overall efficacy [144]. Currently, phage therapy is in its infancy. The current


20

strategy to have phage therapy readily available is development of phage banks. Such banks

are collections of previously characterised phage isolates, which are available as phage stocks,

for direct matching to a specific recently isolated target bacterium and application in cocktails

[160, 181]. Few places around the world have such bank available and one of the biggest phage

therapy centers is located at the Eliava Institute, Georgia [182, 183].

However, despite thorough phage characterisation, in vitro phage performance does not always

match experimental outcomes observed in vivo [157, 181, 184]. The reasons for this may be

diverse. Phages may not adsorb with the same efficiency in vivo due to differences in the

chemical composition of the adsorption conditions [185]. The phage latent periods might be

extended, or the burst sizes might be smaller in vivo, thereby slowing the magnitude or rates of

phage population growth. This might be of a concern especially for lytic infections, where

phages are produced [186]. Notably, phages in vitro are often cultured with bacteria under

somewhat optimal growth conditions, which could differ substantially from in vivo and/or in

situ circumstances. Finally, the target bacteria can differ considerably from the bacterial hosts

against which the phages may have been characterised in vitro [157, 184]. It is clear that

increased knowledge on the host-pathogen interaction is necessary as well as the PK and PD

behaviour of phages is deciphered for a more certain and reliable in vitro and/or in vivo outcome

and hereby successful phage therapy application [160, 187, 188].

Apart from the clinical hurdles, there are also regulatory problems that represent a significant

barrier for the implementation of phage therapy in modern medicine. Unlike the well-

established path to approval for antibiotics, the path for phage therapeutics is currently under

development. The main challenges are the traditional large-scale clinical trials that should be

in accordance with official guidelines and the Good Manufacturing Practices (GMP) in the

production of phage cocktails. Usually, these procedures are very expensive and take several

years, while for phage therapy, for each infection, there may be the need for using another

cocktail [152]. In Belgium, however, a group of researchers recently worked with regulatory

authorities to successfully set up a new regulatory framework for phage therapy [189]. This

new framework classifies phages not as drugs but as active pharmaceutical ingredients, thus

exempting them from clinical trial requirements and allowing them to be administered by

pharmacists on a per-patient basis upon medical prescription. Even though progress has been

made towards overcoming some of the hurdles associated with phage therapy, in order for

phage therapy to gain widespread acceptance or worldwide application profound interest from

big pharmaceutical companies and funding bodies is still needed [152].


21

An alternative to some of the problems described above could be the combined therapy of

phages and antibiotics that take advantage of each treatment’s differing strengths constituting

an ideal synergistic approach [132, 152, 190]. The mechanism of this relies on phages that use

an antibiotic efflux pump to infect the bacterium. This may select against the expression of the

pump, rendering the bacteria more sensitive to antibiotics that were previously pumped out.

This interaction selects for phage-resistant variants, however, as they become more sensitive

to antibiotics, the combined therapy is still effective in inhibiting/killing the target bacteria.

This type of combined therapy has been shown to have an increased effect on several bacterial

species, including E. coli in broiler chickens, compared to when used separately [132, 191,

192].

1.3 Phage-host interactions

Phage-bacterium co-evolution is an important driving force for the ecology and evolution of

microbial communities [193]. However, the nature of some interactions within phage

populations or between different phages and bacteria is only now becoming clear and we are

just starting to understand the complexity of these interactions [194–196]. From an

evolutionary point of view, interactions can be classified as parasitic, predatory, cheating,

mutualistic, or altruistic depending on the system [194, 197, 198]. This classification depends

by large on the life cycle of the phage, including determinants that play a role in the phage host

range and bacterial defence mechanisms [43, 50]. Phages are considered to be parasites when

they exploit bacterial cells for their survival and replication. In the lytic life cycle, when phages

infect and kill their infected host cell(s) they are considered predatory, likewise, they shape the

bacterial population dynamics and may assist in their long-term evolution through generalised

transduction [43]. As phages replicate inside a bacterial cell, pools of public goods (enzymes,

capsid, proteins etc.) are produced. Phage cheaters can emerge and do not contribute to the

production of common goods or consume the goods at a higher rate than the ancestral phages

[194]. In response to phage predation, infected bacterial cells can altruistically arrest their

growth trapping immature phages inside the infected bacterial cell to protect the overall

bacterial population [199]. Also, infected bacterial cells may commit altruistic suicide to halt

phage replication. In the lysogenic life cycle, phages can stably integrate into the host cell

genome or stay as a plasmid inside the cell and may confer lysogenic conversion. Some of

these phages may cause an increased host fitness and diversity as well as function as a survival

strategy for both phage and their host and as such interact mutualistically [43, 194, 198].


22

Interactions between phages and their host(s) have profound effects on biological processes,

prokaryotic metabolism, and diversity and composition of microbial communities [50, 193,

200]. Thus, understanding these interactions not only provides new insights into phage biology

and evolution, the use of phages in genetic engineering and other application(s) but most

importantly may lead to advances in the development of phage therapies [50, 65].

1.3.1 Population growth dynamics

Population dynamics is the study of how and why the population of one or more species

changes in size and structure over time [201]. Accordingly, when studying phage population

growth dynamics, parameters such as phage attachment and adsorption rate, burst size (the

number of phages produced per infected bacterium), latent period (the time period between

adsorption and cell lysis), and life cycle as well as bacterial growth rate and defence

mechanisms affect the dynamics [202, 203]. Phage resistance can occur through various

mechanism in populations of phage and bacteria (see section 1.3.2 ), resulting in partial or

complete resistance, and can differ in the extent of the physiological cost associated with

resistance, and in whether the mutation can be countered by a mutation in the infecting phage.

These important differences determine the effect of the phage infection on the population

dynamics and may have significant consequences for the resulting structure of the microbial

community [204, 205]. Phages can subvert the host’s cellular processes to optimise the

intracellular environment for the phage replication. This is achieved by specific, often toxic,

protein-protein interactions that occur early during phage infection, influencing the

intracellular molecular interactions [206, 207].

Phages may evolve to infect multiple hosts. Such extended host range properties can give rise

to an “arms race” between resistance mutations in the bacterium and the changing host range

[208]. When examining experimental phage communities, however, there seems to be an

asymmetry in the arms race in favour of the bacteria, as some resistance mutations cannot be

countered by host range mutations [204, 209]. However, in natural settings, the arms race

dynamics seems different. The bacteria are more resistant to their contemporary phages than

to ancestor phages, allowing fluctuating selection dynamics and continuous cycles of co -

evolution [205, 210]. This difference in dynamics may be due to the additional biotic and

abiotic selection pressures that are found outside of controlled laboratory conditions [197, 211].


23

Several mathematical models have been developed to predict and explain the behaviour and

dynamics of phage and bacteria populations based on fundamental phage-bacteria biological

parameters [202, 212–214]. Most often, such models are validated using in vitro data obtained

from phage-interaction studies. While no single model to date has been able to capture all

aspects of the complex in vivo interaction between phage and host, together, suitable models

can be selected to predict and explain basic behaviours of selected population dynamics in a

given environment [202, 212, 214]. Microbial model communities have been shown to be ideal

to provide insights into complex microbial community interactions [204]. One important

advantage of phage–bacteria systems is that the initial complexity of a community is

controllable. The complexity can be reduced to a minimum (one phage and one bacterium) and

then increased gradually to examine its effects on population dynamics, community properties,

and evolutionary change. Moreover, both environmental and genetic parameters can relatively

easily be manipulated [204, 215].

When studying phage population dynamics in an animal-associated microbial environment,

one should not only consider the interaction between the phage and the host bacterium, but also

the interplay with the environment within the animal host. The host environment includes a

direct influence of digestive enzymes as well as the influence of non-enzymatic secreted

compounds, such as bile salts, which have been shown to inhibit phage adsorption and

components of the eukaryotic hosts immunity [40, 194, 216]. On the other hand, phages can

have profound effects on the outcome of bacterial infection by modulating the immune

responses of the animal host, either indirectly via effects on the eukaryotic microbiome or

directly, often in anti-inflammatory ways. Phages can modulate the innate immunity of the

mammalian and avian host via the stimulation of phagocytosis and cytokine response, as well

as impact on the adaptive immunity via stimulating the antibody production [173, 217, 218].

Essential knowledge on phage-host interactions may be obtained using mathematical model

outputs using in vitro data from controlled laboratory conditions, followed by in vivo

verification in more complex environments [187]. Subsequent model refinement could be

applied if the experimental data do not reflect the simulated ones [202]. Gaining an

understanding of the phage-bacterium interactions and population dynamics in natural

environments seem essential for future successful phage therapy application as well as

exploiting the full potential of phages for our benefits [219].


24

1.3.2 Bacterial phage resistance

Faced with a strong selection pressure, bacteria can evolve resistance to a phage infection,

either complete or partial, through various mechanisms. These include among others

spontaneous mutations, innate immune systems, including restriction-modification (R-M)

systems, abortive infection (Abi) mechanisms, bacteriophage exclusion (BREX) systems, and

adaptive immunity via the Clustered Regularly Interspaced Short Palindromic Repeats

(CRISPR)-Cas (CRISPR-associated proteins) systems [175, 176, 208, 220–222]. These

antiviral mechanisms can be used to target different steps of the phage life cycle, including

inhibition of phage adsorption and blockage of phage DNA injection, replication of the phage,

or lysis of the bacterium.

Both phage resistance and phage-bacterial co-evolution are mainly driven by spontaneous

mutations [193, 204]. The resistance is largely affecting the adhesin-receptor binding, through

mutation of the receptor or loss of the receptor. This adhesin-receptor binding is highly specific.

The main phage adhesin is the phage tail fiber [223, 224]. The host cell receptors (reviewed in

[224, 225]) of tailed coliphages are surface structures, of which the most often involved surface

structures are outer membrane proteins (OMPs): FhuA (involved in iron uptake, previously

called TonA), OmpC (involved in iron transmembrane transport), OmpF (involved in iron

transmembrane transport), FadL (involved in translocation of long-chain fatty acids across the

membrane), BtuB (involved transports vitamin B12 across the membrane), LamB (involved in

the transport of maltose and maltodextrins), NfrA (required for irreversible adsorption of N4-

like phages), TolC (involved in efflux of antibiotics and other toxic compounds from the cell)

and the TonB protein (involved in the transduction of energy from the cytoplasmic membrane

to the Omps). Other surface structures involved are the O-antigen (part of the LPS of Gram-

negative bacteria), the LPS core, and the pilus (colonisation factor) [226, 227]. In some cases

however, resistance through mutations may lead to an extended host range of the phage [228,

229]. This has been shown for both T2 phages that attach to OmpF, LPS and/or FadL and for

T7 phages that attach to the LPS core. While phage T7 and T4 are phylogenetically unrelated,

both bind to the LPS core, and mutations in the LPS that confer resistance to T4 often gives

cross-resistance to T7. However, LPS mutations that confer cross-resistance do not select for

extended T4 host range. Thus, utilising the same receptor does not guarantee same type of

dynamics [204, 226]. The mutations in the host, giving resistance to the T4 phage and confer

cross‐resistance to phage T7, tend to have a larger fitness cost than the mutations giving only

resistance to T4. This is because the mutation(s), giving cross-resistance, occur deeper in the


25

LPS (the initial binding site for T7 is deeper in the LPS core than is the T4 site) and as a

consequence have a greater effects on the E. coli physiology [204].

Different phage-host dynamics can arise depending on the cost of the mutation. Coliphage T5

has two sets of tail fibers which bind to the O-antigen and/or the FhuA protein [226]. E. coli

resistance to coliphage T5 happens without any fitness costs (under laboratory experimental

settings). The mutations occurring in T5 to counter these resistant bacteria do not confer host‐

range changes, and as a consequence, phage T5 rapidly goes extinct in experimental bacterial

communities when resistance arises [204]. Phage resistance development has also been shown

to influence bacterial virulence. Phage-resistant bacteria may become less virulent in case of

mutations in surface virulence factors, such as LPS, though it depends on which part of the

LPS is affected [156].

In addition to receptor mutations, mechanisms that prevent phage adsorption can be divided

into three categories: blocking of phage receptors, production of extracellular matrix and

presence of competitive inhibitors (Figure 5).

Figure 5 | Defence strategies used by bacteria to prevent phage adsorption. A) Bacteria can produce proteins that

mask the phage receptor. B) Phage adsorption can be prevented by the production of EPS, but some phages

overcome the EPS layer by producing enzymes (lyase or hydrolase) that cleave EPS. C) Bacteria can produce

competitive inhibitors that bind to the phage receptor and reduce or prevent phage adsorption [adapted from

reference [175]].


26

The first category is related to masking the phage receptor(s) through production of masking

proteins that block the access to the receptor from the phage. The second category also includes

hindering access of the phage, but through the production of extracellular matrix of

exopolysaccharides (EPSs) that provides a physical barrier between phages and their receptors

on the host cell surface. However, some phages have evolved to either utilise these extracellular

polymers as receptors or to degrade them. The third category involves competition between

molecules that bind to the same receptors. By procuring competitive inhibitors that bind to the

phage attachment site, the bacterium renders these receptors unavailable for the phage(s). Also,

when the phage receptors play important roles in bacterial metabolism, such as substrate intake,

molecules are binding to the receptor as part of the normal cell activity and might block the

access of the phage [175, 176, 225].

R-M systems are widespread innate defence systems in bacteria [230, 231]. They prevent entry

of foreign DNA into the cell and comprise two contrasting enzymatic activities: a restriction

endonuclease (REase) and a methyltransferase (MTase). The REase recognizes and cleaves

foreign phage DNA sequences at specific sites, while MTase activity ensures discrimination

between host and foreign DNA. As such, these systems may cause phage resistance by cutting

the phage DNA and likewise block the intercellular phage development [40, 231]. Some

phages, however, have evolved several strategies to evade these R-M systems (reviewed by

[232]). One strategy is to select against specific restriction sites. Phages that possess fewer

restriction sites in their genomes are less prone to DNA cleavage by the host REases. A second

strategy includes modification and change of the orientation of restriction-recognition sites to

avoid cleavage by the host REases. A third strategy is for the phage to synthesise proteins that

prevent cleavage by masking the restriction sites. Also, some coliphages while injecting their

own DNA into the host cell, also co-inject host-genome-binding proteins that mask the

restriction sites. A fourth strategy is to disrupt the structural conformation of the REase-MTase

complex. A fifth strategy is for the phage to code for proteins that directly inhibit REase.

Finally, many phages encode their own MTases, which they have acquired from their bacterial

host(s). These enzymes enable phage self -methylation of the DNA hereby protecting against

host restriction enzymes. In turn, some bacteria can evolve to encode modification-specific

endonucleases to restrict and specifically counteract these adapted phages and their

modifications, resulting in a co-evolutionary arms race [231, 233].

Abi systems constitute another mechanism of the bacterial innate immunity. These systems

function as “suicide” systems and induce an altruistic death (or dormancy) of the bacterial cell


27

upon phage infection, and thus ensure no multiplication of the phage [175, 232, 234]. Abi

systems are diverse and can act at any stage of phage replication cycle. They often consist of a

single protein or protein complex encoded by mobile genetic elements, including prophages

and plasmids [232]. Abi can be mediated by toxin-antitoxin (TA) systems encoded by the

bacterium. The TA systems are composed of a toxin, which targets essential cellular process,

such as DNA replication and translation, and a neutralising antitoxin that inhibits the toxin

during normal bacterial growth [234, 235]. Compared to their toxins, antitoxins are more labile

and degrade more rapidly. As a consequence, when stress is encountered (e.g. during phage

infection) and the production of both components is inhibited, the antitoxin is preferentially

degraded, allowing the toxin to induce either bacterial dormancy or cell death [236]. Phages

can evade Abi systems effect by a variety of mechanisms. First, phages can produce

spontaneous "escape" mutants, often conferred by mutation in a single gene (differs between

phages) [237, 238]. Such mutation may hinder depolarisation of the bacterial membrane or

prevent the activation of the Abi system [232]. Second, some virulent phages can become

resistant to specific Abi systems by recombining with the genome of a res ident prophage.

However, the exact mode of action of this evasion strategy remains unclear. Third, phages can

hijack and/or produce antitoxins that neutralises the bacterial toxin [124, 232, 239]. Finally,

some phages are able to produce small RNAs that act as a molecular mimic of the antitoxin

RNA (pseudo-antitoxin), leading to toxin inhibition and avoiding cell death leading to

continued phage replication [240, 241].

In addition to the innate immune systems, CRISPR-Cas systems are found among ~36% of

bacteria and confer a sequence specific adaptive immunity against invading foreign DNA,

including phages [242]. The CRISPR loci consist of arrays of short direct repeats of ~30 base

pairs (bp), separated by similar sized highly variable spacer sequences, derived from invading

nucleic acids, and associated cas genes, encoding Cas proteins [220, 243]. CRISPR-Cas system

immunity involves three main stages (Figure 6). First, the adaptation (or acquisition), where

short DNA fragments from newly encountered foreign genomes are incorporated into the

CRISPR loci as a new unique spacer by the Cas proteins. Second, the expression, whe re the

CRISPR loci is transcribed from a leader sequence upstream of the loci and processed into

small guide CRISPR RNAs (crRNAs). Third, the interference, whereby Cas protein(s) form

complexes with the crRNAs. Using the crRNAs as guides, the crRNA-Cas protein (crRNP)

complexes specifically recognise, bind, and degrade complementary foreign nucleic acids, and

as such kill the phage [244–248].


28

Figure 6 | Simplified overview of bacterial CRISPR-Cas immunity. CRISPR-Cas immunity is mediated through

three steps: (1) Adaptation. A small fragment of the invader phage DNA is acquired and integrated into the

bacterial host CRISPR loci near a “leader” sequence (L). The CRISPR loci contains copies of short direct repeats

(black) that separates the invader phage-derived sequences (coloured boxes). Cas genes (blue) encoding the

protein components of the systems are typically located adjacent to the CRISPR loci. (2) Expression. The

CRISPR loci is transcribed and processed into multiple individual short crRNA molecules. (3) Interference. The

crRNA associate with Cas protein(s) to from a crRNP complex. The crRNP specifically recognizes foreign DNA

(or RNA) via base-paring of the crRNA and cleaves (scissors) in the region of hybridisation. Modified from [243].

In the phage-bacterium arms race, phages utilise different strategies to circumvent the CRISPR-

Cas defence [249, 250]. One strategy includes the acquisition of genetic mutation(s) in the

sequence targeted by the crRNA, making the phage unrecognisable, and consequently,

undetectable from the existing CRSIPR spacers [247]. Another strategy utilised by some

phages is the production of anti-CRISPR (Acr) proteins, which specifically inhibit crRNP

complex DNA binding or interfere with the Cas nucleases, hereby avoiding the degradation of

their phage DNA [249, 251, 252]. Initially, Acr genes were detected mainly in temperate

phages. And only recently in virulent (strictly lytic) phages [253]. Moreover, recent studies

have shown that phages have the abilities to cooperate to overcome CRISPR-Cas immunity

and ensure successful phage infection [250, 254–256]. While initial phage infection may fail


29

due to degradation by the host CRISPR-Cas system(s), Acr production prior to the phage

degradation leaves the cell immunosuppressed. If the cell is re-infected by other phages, Acr

proteins from the initial phage infection increase the likelihood of subsequent successful phage

infection. Moreover, Borges et al. (2018) demonstrated that complete CRISPR-Cas

inactivation by Acr proteins is challenging, and that the concentration of Acr required for the

inactivation and successful infection is contributed by multiple phages. Accordingly, these

initially failed phage infections represent a form of altruism within phage populations: phages

that initiate failed infections suffer a cost in suppressing CRISPR-Cas immunity while kin

phages benefit by initiation successful infection.

Active defence systems, such as R-M enzymes and CRISPR-Cas adaptive immunity, are

widely distributed among bacteria and confer a specific defence against phage infection.

However, the maintenance of such systems has its own fitness costs [231, 257]. For R-M

systems, bacterial fitness cost associated with the production of enzymes involved in the

restriction of foreign DNA exists, while expression of Cas proteins is particularly costly and

associated with lower competitive abilities [258].

It remains unclear what is the effect of the co-evolutionary arms race between phage and host

and the possible emergence of phage-resistant bacteria variants [176]. This has major

implications for the use of phages in therapy as the speed of the emergence of resistance may

influence the outcome. In vitro, resistance has been shown to develop in the timespan of hours

to days [147]. Whether the evolution of phage resistance in vitro is comparable to in vivo and/or

in situ conditions where bacteria may be replicating more slowly and challenged with a greater

set of environmental conditions needs further investigation and most likely will depend on the

type of phage, pathogen and their interaction(s) [176]. Accordingly, in vitro resistance selection

experiments might not fully account for the complexity of the phage-host interaction and co-

evolution dynamics [133, 147]. Nevertheless, in vitro phage resistance characterisation might

be an important first step in assessing the relative likelihood of emerging phage-resistant

bacterial populations, the most likely phenotype(s) of resistant mutants, as well as the effect of

combinations of phages on the resistance development [214, 259].


30

1.4 Avian pathogenic Escherichia coli (APEC)

1.4.1 Diseases, transmission, and reservoirs

Escherichia coli (E. coli) belongs to the Enterobacteriaceae family, and is a Gram-negative,

facultative anaerobe, motile rod-shaped bacterium with approximately 0.5 µm in diameter and

1-3 µm in length [260, 261]. E. coli is a ubiquitous bacterium widely distributed in various

environments and the global population size has been estimated to be 10 20 [262]. The

gastrointestinal (GI) tract of humans and animals is considered to be the primary habitat of E.

coli, whereas the external environment serve as a secondary habitat, in which the bacteria are

excreted (e.g. water, soil, and sediments) [263, 264]. Although most E. coli strains are harmless

or beneficial commensals of the GI tract, some strains are highly virulent pathogens that can

cause a variety of infections [265–267]. The majority of pathogenic E. coli strains are

considered opportunistic pathogens as they exist most of the time harmlessly as commensals

of the microflora in a wide spectrum of hosts, and only cause infections under certain

conditions, including weakened immune system of the host and presence of bacterial stress

factors [267, 268].

Pathogenic E. coli can be classified into two main groups according to the infection site and

clinical outcome: intestinal or diarrheagenic E. coli, if they cause infections inside the intestine,

and extraintestinal E. coli (ExPEC), if they cause infections in extra-intestinal sites [265].

ExPEC strains that cause disease in poultry are characterised as avian pa thogenic E. coli

(APEC). APEC is recognised as one of the most important bacterial pathogens of poultry and

other avian species. APEC causes a large range of localised or systemic extra -intestinal

infections, which collectively are referred to as colibacillosis [269]. These infections can result

in morbidity and mortality, and hereby, significant economic losses to the poultry industry

globally [268–270]. Healthy birds with a normal immune system generally do not develop

disease. Thus, the different infection types of colibacillosis and corresponding pathological

manifestations depend on the routes of entry of APEC, the tissue(s) affected, virulence

properties of the strain, host status (e.g. species, age, type of production, and health status), and

the presence of predisposing factors [271]. Such factors include damaged mucosal epithelial

barriers (e.g. skin wounds and mucosal damage from viral, bacterial, and parasitic infections),

impaired or suppressed immune system (e.g. due to viral infections or nutritional deficiencies),

or inappropriate husbandry practices (e.g. contaminated environments and abnormal stress).

Accordingly, colibacillosis in poultry often occur as a secondarily localised or systemic disease


31

when the host defences have been impaired [268]. In Belgium, APEC infections were identified

as a major factor in poultry disease, and the incidence of APEC infections in layers, breeders,

and broilers has been shown to be 38.6%, 26.9%, and 17.7%, respectively [272]. The primary

reservoir of APEC strains is the intestinal tract of poultry. Poultry may carry up to 106 colony

forming units (CFU) of E. coli per gram of faeces, of which an estimate of 10-15% belong to

potentially APEC serogroups [268, 273]. Excretion to the environment allow for bird-to-bird

transmission of E. coli strains via the faecal-oral route, and the bacteria can survive in dust of

poultry houses reaching concentrations of ~105-106 CFU/g [268, 269]. Poultry litter may reach

levels as high as 108 CFU/g [274].

1.4.2 Virulence factors

APEC possesses and utilises various virulence factors, including adhesins, iron acquisition

systems, protectins, and toxins to cause infection in poultry (reviewed in [253]) (Table 2).

These factors are essential for APEC to adhere, invade, evade the host immune responses,

colonise, and cause infection in extraintestinal sites [267, 269]. Adherence of the bacterium to

epithelial surfaces of the host is an essential step in the APEC infection required for host

colonisation. The adherence is primary mediated by pili/fimbriae located on the outer

membrane of most strains [275]. The P pili are thought to play a role by means of their PapG

adhesin, which is found in three variants: PapGI, PapGII, and PapGIII, encoded by the three

alleles of the corresponding gene, papG. Especially papGII is more likely to be found among

APEC isolates compared to avian faecal E. coli isolates [271, 276, 277]. Besides, the outer

membrane protease OmpT, encoded by the ompT gene, is thought to participate in the adhesion

of APEC as well as antibiotic resistance through peptide cleavage [278, 279]. Pathogenic E.

coli have been shown to possess high prevalence of genes, such as sitA, iutA, fyuA, irp2, ireA

and iroN, coding for iron acquisition systems [275, 280]. Once the bacteria have successfully

colonised the host, these systems may contribute to the APEC virulence by facilitating

acquisition of iron, which is essential for bacterial growth, proliferation, and protection against

environmental stresses. Moreover these systems may promote adaptation to colonise and infect

sites where iron is depleted by antibacterial host defences [269, 275, 276, 281]. Protectins

provide protection of the bacteria against the host immune system as well as unfavourable

conditions, and include, among others, bacterial capsule, OMPs, and LPS components [271,

275]. Multiple toxin types have been reported in APEC and assist in the ability of the bacteria

to invade and cause damage to the tissues [269].


32

Table 2 | Prevalent virulence-associated traits of avian pathogenic E. coli (APEC)

Category Gene(s) Description Ref

Adhesins pap P fimbriae [271]

ompT Outer membrane protein T (protease) [278]

Toxins hlyF* Hemolysin F [282]

tsh Temperature-sensitive haemagglutinin auto-transporter [283]

Iron acquisition

systems

fyuA Yersiniabactin (siderophore) synthesis and receptor [283]

ireA Iron regulated element (catecholate siderophore receptor) [284]

iroN* Catecholate siderophore receptor [285]

irp2 Yersiniabactin (siderophore) synthesis and receptor [286]

iutA* Aerobactin (siderophore) synthesis and receptor [285]

sitABCD Iron and manganese transport system [281]

Protectins

cvaC* Colicin V (ColV) [276]

iss* Increased serum survival, complement resistance [285]

neu Capsular polysialic acid biosynthesis [287]

wzy O-antigen polymerase [288]

Other fliC Flagellar antigen H7 [276]

* Plasmid-encoded. Modified from [269, 289].

The majority of APEC strains are often characterised by the presence of large conjugative

plasmids, harbouring a number of virulence genes [276, 290, 291]. Several of these plasmid-

encoded genes, such as the ColV-type plasmid-encoded genes cvaC, and iss provide a

competitive advantage for nutrient acquisition through colicin production and increased serum

survival of the bacterium, respectively [292–299]. Also frequently located on a ColV-type

plasmid is the temperature-sensitive hemagglutinin gene, tsh, which encodes the

autotransporter Tsh protein that play a role in the pathogenicity of E. coli in the early stages of

infection [300]. It has been shown that transferring APEC ColV-type plasmids into

avirulent/commensal E. coli strains enhances their ability to colonise and infect hosts in

experimental models, supporting a role of plasmid-encoded genes in the pathogenicity of E.

coli [301]. Compared to avian faecal E. coli isolates, several virulence genes have more

frequently been observed in APEC, including the five plasmid-encoded genes (iutA, hlyF, iss,

iroN, and ompT), which have been identified as predictors of APEC virulence [276, 285, 293,

302, 303].


33

1.4.3 Strain typing and population genetics

Since the 1940s, bacteria have been serotyped based on their surface antigens: the somatic O-

LPS antigen, the flagellar H-antigen, and the capsular K-antigen, which define serogroup (O

antigen only) or serotype (O, H, and K antigens) [304]. More than 185 O-, 60 H-, and 80 K-

antigens have been recognised, and variations of various different serotypes combinations have

been identified [305, 306]. Multiple APEC serogroups have been associated with colibacillosis

cases, however, the three O serogroups O1, O2, and O78 constitute more than 80% of the cases

[269, 307, 308]. Other APEC infection-associated serogroups include O8, O18, O35, O109,

and O115 [276, 309].

In the late 1990s, MLST emerged as a powerful method for analysis of bacterial population

genetics and phylogenetic relationship [310]. The principle of MLST is to identity nucleotide

sequence variation in 400-500 bp of seven conserved housekeeping genes. The genetic

variation is characterised by sequencing the fragments of housekeeping genes. An identifier

number is given to each unique sequence (allele) recognised for each locus, and a sequence

type (ST) is given to the specific combination of alleles at all loci. Together, this loci

combination is used to generate a DNA fingerprint of the bacterial isolate. The most widely

used MLST scheme for typing of E. coli strains is Achtman’s scheme

(https://pubmlst.org/data/), which uses seven housekeeping genes for strain discrimination

[311]. Recently, APEC (isolates belonging to the ST95 and ST131) has been p resented as a

potential foodborne zoonotic pathogen and as such, is a pathogen of importance to the poultry

industry as well as the public health [269].

In recent years, high-throughput sequencing technologies have made it possible to understand

the population genomics at the single gene level as well as the whole-genome level. WGS data

has allowed for the development of high-resolution typing methods that circumvents previous

typing challenges and enable comparisons of inter- and intraspecies bacterial genomes.

Besides, WGS typing approaches, such as discovery of single nucleotide polymorphisms

(SNPs), offer greater phylogenetic resolution as well as genetic markers to study evolution

[312–314].

1.4.4 Current strategies to prevent and control APEC

The commercial poultry industry depends on cost efficiently raising birds in large quantities

[315]. The current prevention and control of APEC infections in poultry relies on biosecurity

https://pubmlst.org/data/


34

measures to reduce the pathogen load, management of environmental stressors, application of

feed additives, vaccination strategies, and antimicrobial treatment [269, 315, 316].

Prevention and control of APEC infections (colibacillosis) in poultry depends largely on

identifying and eliminating the predisposing cause(s) of the disease outbreak [289]. Keeping

strict biosecurity measures (such as segregation, traffic control, cleaning, and disinfection)

helps to prevent a large proportion of harmful bacteria and viruses from entering and spread

within the poultry facilities [269, 315, 317], however, it is necessary to make these measures

practical, enforceable and cost effective [269, 315]. Environmental factors should be well

monitored, controlled, and adjusted. Proper ventilation as well as maintaining optimum

temperature, humidity, and bird density will keep ammonia and dust in poultry facilities low

and thereby reducing the risk of APEC infections [269, 289].

A variety of different feed additives are commonly used to control E. coli in poultry, including

prebiotics, probiotics, enzymes, acidifiers, vitamins, immune enhances, and other

antimicrobials [269, 289].These may enhance wound healing or promote intestinal health, and

thus, directly or indirectly help preventing infections [315].

Vaccination is commonly used in the control of infections. While some vaccines are given to

protect the individual bird against disease, others are given to increase maternal immunity

[315]. Various E. coli vaccines have been developed to prevent colibacillosis (to varying effect)

[269, 317]. These include inactivated, live-attenuated, recombinant and subunit vaccines [289].

Despite multiple vaccine candidates with proven in vivo efficacy, only two vaccines (live-

attenuated APEC O78 ΔaroA Poulvac® and inactivated Nobilis® comprising F11 fimbrial and

FT flagellar antigens) are currently commercially available for use in chickens [269, 316]. The

major drawback of these vaccines is the lack of protection against heterogenous APEC

infections, comprised of multiple serotypes [289]. In large-scale poultry productions, efficient

and economic application of vaccines is a challenge. The most commonly used application

techniques include in ovo injection, subcutaneous or intramuscular injection, spray, intraocular

or nasal drop, and through the drinking water [315]. Application technique depends on the

vaccine used [269, 315]. The ideal APEC vaccine should be able to confer cross-protection

against multiple APEC serotypes and be deliverable by mass-vaccination methods, such as in

ovo, oral (feed or water), or spray routes [269].

Antimicrobials are commonly used to treat colibacillosis [269]. When thousands of birds are

grouped together in commercial facilities, segregation and treatment of individuals is


35

impractical and labour-intensive, so metaphylactic use is applied [315]. Accordingly, the

majority of antibiotics used in poultry production are administered orally, mixed into feed or

drinking water, using automated systems [318, 319]. As sick birds often have little or no

appetite and are unable to compete for feed, water medication may be preferred over feed

medication, as sick animals will still frequently drink [289, 315]. While antimicrobial treatment

may not be completely successful in curing sick animals, it may hold the disease in check until

cleared by the immune system of the host [269]. However, the increasing resistance of APEC

strains to multiple antibiotics limits the use of antimicrobials [269]. Moreover, restrictions on

antimicrobial use in poultry have been imposed by regulatory and public concern [289]. The

United States of America and European Union (EU) have employed strategies to restrict the

non-therapeutic use (for growth promotion) of antibiotics in food–animal production and to

limit the therapeutic use (for treatment of sick animals with a diagnosis) of medically important

antibiotics [320].

1.4.5 Phage therapy against APEC infections

Phage therapy has proven to be a potential therapeutic against APEC. However, while several

studies have described efficacies, others have not been able to show efficacy in avian species

[321–330]. Moreover, comparing studies is challenging as different infection models and

treatment protocols were used.

When comparing the infection models used, great variation can be observed between different

variables. 1) The age of the birds (range between 1-day-old and 10-weeks-old). 2) The route of

infection, including direct injection to the air sac, intratracheal (IT) inoculation, oral

inoculation, intramuscular (IM) inoculation, intracranial (IC) injection, or the use of naturally

infected birds. 3) The challenge dose of E. coli used in the infection models (range between

103 and 108 CFU/ml). 4) The bacterial strains used, including APEC and ExPEC isolates with

serogroup O1, O2, O18, O78, O86, O126 or unknown. As for the infection model, several

parameters differ in the treatment protocols used. 1) Prevention and/or therapeutical use has

been reported for respiratory infection/colibacillosis, septicaemia and meningitis. 2) The routes

of administration of phage treatment include air sac inoculation, drinking water, coarse spray

(on chicken or litter), fine spray, orally syringe combined with spray, IT inoculation, IM

inoculation, or IC injection. 3) The dose of administration and formulation. The concentration

of phage(s) applied ranges between 102 and 109 PFU/ml. Most studies include only a single


36

phage in the formulation. Few studies have investigated the difference in efficacy between a

single phage alone and a cocktail of phages. 4) The treatment models include great differences

when comparing time and type of phage administration. Single-dose administration included

prophylactic administration (between one and five days prior to E. coli challenge),

simultaneous administration of phage and E. coli, and therapeutic administration (eight hours

post E. coli challenge). Multiple-dose administration included once a day for seven days prior

to challenge as well as once a day for seven days both prior and post E. coli challenge.

Several variables have been identified that influence differences in outcome (Table 3 ). 1)

Concentration and timing of administered phage(s) are known to effect the success rate of

phage therapy [331]. Independent of infection and treatment model (aerosol spray, orally,

thoracic air sac inoculation, IM inoculation, or IC inoculation), higher phage dosages have

shown more effective (decreased mortality and/or morbidity) [321, 325, 329, 330]. However,

timing of the phage administration has been shown to be of extremely importance for a

successful outcome, as administration of the same high phage titer on different timepoint

(before and/or after bacterial inoculation) resulted in significant differences in outcomes [328,

331]. 2) Phage therapy success rates also depend on the site and route of administration of the

phages. Oliveira et al. (2010), confirmed that colibacillosis-induced morbidity and mortality

may be significantly reduced by spraying of housing systems with phage cocktails as well as

oral administration of phages. However, Huff et al. (2002) found that the best results were

obtained through direct application of the phages on the site of infection. Likewise, Huff et al.

(2013) did not obtain a reduced mortality when phages were administrated via spray while IT

administration was more successful, which was attributed to the fact that only direct

administration could bring high enough phage titers at the site of infection. Similarly, IM

delivery during septicaemia was very effective. On the contrary, high phage titers present in

drinking water did not prevent infection after severe thoracic air sac challenge, as the birds

were not able to absorb enough phages from the water [325]. Thus, while phages targeting

systemic infections may be delivered IM or intravenous, it might be preferable to deliver

phages targeting respiratory infection by inhalation [332]. Moreover, when phages are

administered by spray, the spray type should be taken into considera tion, as smaller droplet

aerosols may penetrate the respiratory tract to a greater extent compared to coarse spray [325].


37

Table 3 | Phage therapy study outcomes based on the body weight and mortality of the birds.

Positive outcomes Negative outcomes

Prevention

• High titer (108 and 109 PFU/ml) phage

in aerosol spray applied

simultaneously with E. coli challenge

significantly decreased mortality [329].

• High titer (108 PFU/ml) phage sprayed

on litter applied simultaneously with

high (108 CFU/ml) oral E. coli

challenge reduced mortality and

morbidity [327].

• High titer (108 PFU/ml) phage mixed

with E. coli prior to challenging birds

via thoracic air sac inoculation

provided complete protection of the

birds [325].

• Low titer (104 PFU/ml) phage mixed

with E. coli prior to challenging birds

via thoracic air sac inoculation

significantly reduced mortality from

85% to % [325].

• High titer (108 PFU/ml) phage

administered IT just before IT E. coli

challenge provided complete

protection [323].

High titer (109 PFU/ml) phage coarse

or fine spray did not protect birds

from IT E. coli challenge that cause

significant decrease in body weight

and significant increase in mortality

[323].

• Phage titer (103-106) in drinking

water had little or no efficacy to

prevent infection when birds were

challenged with large inoculum of E.

coli via thoracic air sac inoculation,

resulting in significant decrease in

body weight and increase in mortality

[325].

Treatment

• High titer (107 PFU/ml) phage cocktail

administered in both drinking water

and fine spray decreased mortality in

large commercial flock naturally

infected [321].

• High titer (106 or 108 PFU/ml) phage

administered IC 8 hours post IC E. coli

challenge significantly reduced

morbidity and mortality [328].

• High titer (109 PFU/ml) phage

administered IM 8 hours post E. coli

challenge via thoracic air sac

significantly reduced the mortality.

Phage cocktail showed more efficient

than single phage [324].

• High titer (108 and 109 PFU/ml)

phage in aerosol spray applied

simultaneously with E. coli challenge

was not an effective treatment as only

low or no titers of phages was

delivered to the blood [329].

• Low titer (102 PFU/ml) phage

administrated IM did not result in any

statistically significant protection

against IM E. coli challenge [328].

IT = Intratracheally, IC = Intracranially, IM = Intramuscularly.


38

3) Phage cocktails have been shown more efficient in treating colibacillosis than a single phage

treatments, probably due to a synergistic effect between the individual phages [324, 333].

However, a cocktail of phages does not necessarily lead to successful application as

demonstrated by Tsonos et al. (2014). Though the cocktail was composed of well-characterised

phages and administered via three different routes (IT, intraoesophageally or via the drinking

water) in high dose, they could not detect a decrease in mortality, lesion scores or weight loss.

Essentially in phage therapy is that the appropriate phage(s) are delivered timely in sufficient

quantities at the site of bacterial infection. Even though this seems to be a relatively easy

criterion to meet, reality show that it is not. As modern poultry production facilities contain

thousands of birds, it would not be feasible to administer an IM treatment of phage during an

outbreak of colibacillosis. If successful, treatment through drinking water would be a practical

method to administer phage in the poultry industry [334].

Despite many promising phage therapy results, phage-based product for treatment of

colibacillosis in poultry are still not available on the market [269, 334]. A great deal of work is

required to select the most efficacious phage(s) against specific pathogenic target bacteria, and

to determine how, when, and how much phage should be administered to provide sufficient

protection to poultry [334, 335]. If we learn practical ways to exploit the phage therapy

potential, phage(s) could provide a valuable alternative or supplement to the use of antibiotics

to prevent and/or control bacterial diseases in poultry [269].


39

References

[1] Gaynes, R. The Discovery of Penicillin—New Insights After More Than 75 Years of Clinical Use.

Emerg. Infect. Dis., 2017, 23 (5), 849–853. https://doi.org/10.3201/eid2305.161556.

[2] Aminov, R. I. A Brief History of the Antibiotic Era: Lessons Learned and Challenges for the Future.

Front. Microbiol., 2010, 1, 134. https://doi.org/10.3389/fmicb.2010.00134.

[3] Gordillo Altamirano, F. L.; Barr, J. J. Phage Therapy in the Postantibiotic Era. Clin. Microbiol. Rev.,

2019, 32 (2), 1–25. https://doi.org/10.1128/CMR.00066-18.

[4] Gould, K. Antibiotics: From Prehistory to the Present Day. J. Antimicrob. Chemother., 2016, 71 (3),

572–575. https://doi.org/10.1093/jac/dkv484.

[5] Hutchings, M. I.; Truman, A. W.; Wilkinson, B. Antibiotics: Past, Present and Future. Curr. Opin.

Microbiol., 2019, 51 (October 2019), 72–80. https://doi.org/10.1016/j.mib.2019.10.008.

[6] Ventola, C. L. The Antibiotic Resistance Crisis: Part 1: Causes and Threats. P T, 2015, 40 (4), 277–

283. https://doi.org/10.24911/IJMDC.51-1549060699.

[7] Davies, J.; Davies, D. Origins and Evolution of Antibiotic Resistance. Microbiol. Mol. Biol. Rev.,

2010, 74 (3), 417–433. https://doi.org/10.1128/mmbr.00016-10.

[8] Silver, L. L. Challenges of Antibacterial Discovery. Clin. Microbiol. Rev., 2011, 24 (1), 71–109.

https://doi.org/10.1128/CMR.00030-10.

[9] World Health Organization (WHO). Antimicrobial Resistance: Global Report on Surveillance;

Geneva, Switzerland, 2014. https://doi.org/10.1007/s13312-014-0374-3.

[10] Ribeiro da Cunha; Fonseca; Calado. Antibiotic Discovery: Where Have We Come from, Where Do

We Go? Antibiotics, 2019, 8 (2), 45. https://doi.org/10.3390/antibiotics8020045.

[11] ReAct Group. No Title https://www.reactgroup.org/toolbox/understand/how-did-we-end-up-

here/few-antibiotics-under-development/.

[12] Centers for Disease Control and Prevention (CDC). Antibiotic Resistance Threats in the United

States, 2019; Atlanta, Georgia, 2019. https://doi.org/10.15620/cdc:82532.

[13] zur Wiesch, P. A.; Kouyos, R.; Engelstädter, J.; Regoes, R. R.; Bonhoeffer, S. Population Biological

Principles of Drug-Resistance Evolution in Infectious Diseases. Lancet Infect. Dis., 2011, 11 (3),

236–247. https://doi.org/10.1016/S1473-3099(10)70264-4.

[14] Rodríguez-Rojas, A.; Rodríguez-Beltrán, J.; Couce, A.; Blázquez, J. Antibiotics and Antibiotic

Resistance: A Bitter Fight against Evolution. Int. J. Med. Microbiol., 2013, 303 (6–7), 293–297.

https://doi.org/10.1016/j.ijmm.2013.02.004.

[15] Goossens, H. Antibiotic Consumption and Link to Resistance. Clin. Microbiol. Infect., 2009, 15

(SUPPL. 3), 12–15. https://doi.org/10.1111/j.1469-0691.2009.02725.x.

[16] Llor, C.; Bjerrum, L. Antimicrobial Resistance: Risk Associated with Antibiotic Overuse and

Initiatives to Reduce the Problem. Ther. Adv. Drug Saf., 2014, 5 (6), 229–241.

https://doi.org/10.1177/2042098614554919.

[17] Holmes, A. H.; Moore, L. S. P.; Sundsfjord, A.; Steinbakk, M.; Regmi, S.; Karkey, A.; Guerin, P. J.;

Piddock, L. J. V. Understanding the Mechanisms and Drivers of Antimicrobial Resistance. Lancet,

2016, 387 (10014), 176–187. https://doi.org/10.1016/S0140-6736(15)00473-0.

[18] von Wintersdorff, C. J. H.; Penders, J.; van Niekerk, J. M.; Mills, N. D.; Majumder, S.; van Alphen,

L. B.; Savelkoul, P. H. M.; Wolffs, P. F. G. Dissemination of Antimicrobial Resistance in Microbial

Ecosystems through Horizontal Gene Transfer. Front. Microbiol., 2016, 7 (Feb), 173.

https://doi.org/10.3389/fmicb.2016.00173.


40

[19] Wozniak, R. A. F.; Waldor, M. K. Integrative and Conjugative Elements : Mosaic Mobile Genetic

Elements Enabling Dynamic Lateral Gene Flow. Nat. Rev. Microbiol., 2010, 8, 552–563.

https://doi.org/10.1038/nrmicro2382.

[20] Skippington, E.; Ragan, M. A. Lateral Genetic Transfer and the Construction of Genetic Exchange

Communities. FEMS Microbiol. Rev., 2011, 35 (5), 707–735. https://doi.org/10.1111/j.1574-

6976.2010.00261.x.

[21] Lopatkin, A. J.; Sysoeva, T. A.; You, L. Dissecting the Ef fects of Antibiotics on Horizontal Gene

Transfer: Analysis Suggests a Critical Role of Selection Dynamics. BioEssays, 2016, 38 (12), 1283–

1292. https://doi.org/10.1002/bies.201600133.

[22] Chen, I.; Christie, P. J.; Dubnau, D. The Ins and Outs of DNA Transfer in Bacteria. Science (80-. ).,

2005, 310 (5753), 1456–1460. https://doi.org/10.1126/science.1114021.

[23] Balcazar, J. L. Bacteriophages as Vehicles for Antibiotic Resistance Genes in the Environment. PLoS

Pathog., 2014, 10 (7), e1004219. https://doi.org/10.1371/journal.ppat.1004219.

[24] Partridge, S. R.; Kwong, S. M.; Firth, N.; Jensen, S. O. Mobile Genetic Elements Associated with

Antimicrobial Resistance. Clin. Microbiol. Rev., 2018, 31 (4), e00088-17.

https://doi.org/10.1128/CMR.00088-17.

[25] Gómez-Gómez, C.; Blanco-Picazo, P.; Brown-Jaque, M.; Quirós, P.; Rodríguez-Rubio, L.; Cerdà-

Cuellar, M.; Muniesa, M. Infectious Phage Particles Packaging Antibiotic Resistance Genes Found

in Meat Products and Chicken Feces. Sci. Rep., 2019, 9 (13281). https://doi.org/10.1038/s41598-

019-49898-0.

[26] Berglund, B. Environmental Dissemination of Antibiotic Resistance Genes and Correlation to

Anthropogenic Contamination with Antibiotics. Infect. Ecol. Epidemiol., 2015, 5 (1), 28564.

https://doi.org/10.3402/iee.v5.28564.

[27] White, A.; Hughes, J. M. Critical Importance of a One Health Approach to Antimicrobial Resistance.

Ecohealth, 2019, 16, 404–409. https://doi.org/10.1007/s10393-019-01415-5.

[28] McEwen, S. A.; Collignon, P. J. Antimicrobial Resistance: A One Health Perspective. Microbiol.

Spectr., 2018, 6 (2), ARBA-0009-2017. https://doi.org/10.1128/microbiolspec.ARBA-0009-2017.

[29] Guardabassi, L.; Butaye, P.; Dockrell, D. H.; Fitzgerald, J. R.; Kuijper, E. J. One Health: A

Multifaceted Concept Combining Diverse Approaches to Prevent and Control Antimicrobial

Resistance. Clin. Microbiol. Infect., 2020, No. xxxx. https://doi.org/10.1016/j.cmi.2020.07.012.

[30] Robinson, T. P.; Bu, D. P.; Carrique-Mas, J.; Fèvre, E. M.; Gilbert, M.; Grace, D.; Hay, S. I.;

Jiwakanon, J.; Kakkar, M.; Kariuki, S.; et al. Antibiotic Resistance Is the Quintessential One Health

Issue. Trans. R. Soc. Trop. Med. Hyg., 2016, 110 (7), 377–380.

https://doi.org/10.1093/trstmh/trw048.

[31] Collignon, P. J.; McEwen, S. A. One Health-Its Importance in Helping to Better Control

Antimicrobial Resistance. Trop. Med. Infect. Dis., 2019, 4 (1), 22.

https://doi.org/10.3390/tropicalmed4010022.

[32] Pires, D. P.; Costa, A. R.; Pinto, G.; Meneses, L.; Azeredo, J. Current Challenges and Future

Opportunities of Phage Therapy. FEMS Microbiol. Rev., 2020, fuaa017, 1–17.

https://doi.org/10.1093/femsre/fuaa017.

[33] Lin, D. M.; Koskella, B.; Lin, H. C. Phage Therapy: An Alternative to Antibiotics in the Age of

Multi-Drug Resistance. World J. Gastrointest. Pharmacol. Ther., 2017, 8 (3), 162–173.

https://doi.org/10.4292/wjgpt.v8.i3.162.

[34] Interagency Coordination Group on Antimicrobial Resistance (IACG). No Time to Wait: Securing

the Future from Drug-Resistant Infections; 2019.

[35] Brives, C.; Pourraz, J. Phage Therapy as a Potential Solution in the Fight against AMR: Obstacles

and Possible Futures. Palgrave Commun., 2020, 6, 100.


41

[36] Twort, F. W. An Investigation on the Nature of Ultra -Microscopic Viruses. Lancet, 1915, 186 (4814),

1241–1243. https://doi.org/10.1016/S0140-6736(01)20383-3.

[37] Summers, W. C. Felix D’Herelle and the Origins of Molecular Biology; Yale University Press: New

Haven, Conn, 1999.

[38] Güemes, A. G. C.; Merry Youle; Cantú, V. A.; Felts, B.; Nulton, J.; Rohwer, F. Viruses as Winners

in the Game of Life. Annu. Rev. Virol., 2016, 3 (1), 197–214. https://doi.org/10.1146/annurev-

virology-100114-054952.

[39] Hatfull, G. F. Dark Matter of the Biosphere: The Amazing World of Bacteriophage Diversity. J.

Virol., 2015, 89 (16), 8107–8110. https://doi.org/10.1128/jvi.01340-15.

[40] Letarov, A.; Kulikov, E. The Bacteriophages in Human- and Animal Body-Associated Microbial

Communities. J. Appl. Microbiol., 2009, 107 (1), 1–13. https://doi.org/10.1111/j.1365-

2672.2009.04143.x.

[41] Batinovic; Wassef; Knowler; Rice; Stanton; Rose; Tucci; Nittami; Vinh; Drummond; et al.

Bacteriophages in Natural and Artificial Environments. Pathogens, 2019, 8 (3), 100.

https://doi.org/10.3390/pathogens8030100.

[42] Abedon, S. T. Phages, Ecology, Evolution. In Bacteriophage Ecology: Population Growth,

Evolution, and Impact of Bacterial Viruses; Abedon, S. T., Ed.; Cambridge University Press:

Cambridge, 2008; pp 1–28. https://doi.org/10.1017/CBO9780511541483.004.

[43] Clokie, M. R. J.; Millard, A. D.; Letarov, A. V; Heaphy, S. Phages in Nature. Bacteriophage, 2011,

1 (1), 31–45. https://doi.org/10.4161/bact.1.1.14942.

[44] Weinbauer, M. G.; Rassoulzadegan, F. Are Viruses Driving Microbial Diversification and Diversity?

Environ. Microbiol., 2004, 6 (1), 1–11. https://doi.org/10.1046/j.1462-2920.2003.00539.x.

[45] de Jonge, P. A.; Nobrega, F. L.; Brouns, S. J. J.; Dutilh, B. E. Molecular and Evolutionary

Determinants of Bacteriophage Host Range. Trends Microbiol., 2019, 27 (1), 51–63.

https://doi.org/10.1016/j.tim.2018.08.006.

[46] Bielke, L.; Higgins, S.; Donoghue, A.; Donoghue, D.; Hargis, B. M. Salmonella Host Range of

Bacteriophages That Infect Multiple Genera. Poult. Sci., 2007, 86 (12), 2536–2540.

https://doi.org/10.3382/ps.2007-00250.

[47] Yu, P.; Mathieu, J.; Li, M.; Dai, Z.; Alvarez, P. J. J. Isolation of Polyvalent Bacteriophages by

Sequential Multiple-Host Approaches. Appl. Environ. Microbiol., 2016, 82 (3), 808–815.

https://doi.org/10.1128/AEM.02382-15.

[48] Grose, J. H.; Casjens, S. R. Understanding the Enormous Diversity of Bacteriophages: The Tailed

Phages That Infect the Bacterial Family Enterobacteriaceae. Virology, 2014, 468, 421–443.

https://doi.org/10.1016/j.virol.2014.08.024.

[49] Ross, A.; Ward, S.; Hyman, P. More Is Better: Selecting for Broad Host Range Bacteriophages.


[50] Díaz-Muñoz, S. L.; Koskella, B. Bacteria –Phage Interactions in Natural Environments. In Advances

in Applied Microbiology; Elsevier Inc., 2014; Vol. 89, pp 135–183. https://doi.org/10.1016/B978-0-

12-800259-9.00004-4.

[51] Hyman, P. Phages for Phage Therapy: Isolation, Characterization, and Host Range Breadth.

Pharmaceuticals, 2019, 12 (1), 35. https://doi.org/10.3390/ph12010035.

[52] Casjens, S. R. Comparative Genomics and Evolution of the Tailed -Bacteriophages. Curr. Opin.

Microbiol., 2005, 8 (4), 451–458. https://doi.org/10.1016/j.mib.2005.06.014.

[53] Scholl, D.; Rogers, S.; Adhya, S.; Merril, C. R. Bacteriophage K1-5 Encodes Two Different Tail

Fiber Proteins, Allowing It To Infect and Replicate on Both K1 and K5 Strains of Escherichia Coli.

J. Virol., 2001, 75 (6), 2509–2515. https://doi.org/10.1128/jvi.75.6.2509-2515.2001.


42

[54] International Committee on Taxonomy of Viruses (ICTV), A. M. K. Virus Taxonomy: Ninth Report

of the International Committee on Taxonomy of Viruses, 1st ed.; King, A. M. ., Adams, M. J.,

Carstens, E. B., Lefkowitz, E. J., Eds.; Elsevier, 2011.

[55] Ackermann, H.-W. 5500 Phages Examined in the Electron Microscope. Arch. Virol., 2007, 152 (2),

227–243. https://doi.org/10.1007/s00705-006-0849-1.

[56] Dion, M. B.; Oechslin, F.; Moineau, S. Phage Diversity, Genomics and Phylogeny. Nat. Rev.

Microbiol., 2020, 12. https://doi.org/10.1038/s41579-019-0311-5.

[57] Krupovic, M.; Turner, D.; Morozova, V.; Dyall, M.; Hanna, S.; Edwards, R.; Dutilh, B. E.; Lehman,

S. M.; Reyes, A.; Baquero, D. P. Bacterial Viruses Subcommittee and Archaeal Viruses

Subcommittee of the ICTV : Update of Taxonomy Changes in 2021. Arch. Virol., 2021, 1–6.

https://doi.org/10.1007/s00705-021-05205-9.

[58] Zrelovs, N.; Dislers, A.; Kazaks, A. Motley Crew: Overview of the Current ly Available Phage

Diversity. Front. Microbiol., 2020, 11. https://doi.org/10.3389/fmicb.2020.579452.

[59] Turner, D.; Kropinski, A. M.; Adriaenssens, E. M. A Roadmap for Genome-Based Phage Taxonomy.

Viruses, 2021, 13 (3), 506. https://doi.org/10.3390/v13030506.

[60] Veesler, D.; Cambillau, C. A Common Evolutionary Origin for Tailed-Bacteriophage Functional

Modules and Bacterial Machineries. Microbiol. Mol. Biol. Rev., 2011, 75 (3), 423–433.

https://doi.org/10.1128/mmbr.00014-11.

[61] Aksyuk, A. A.; Leiman, P. G.; Kurochkina, L. P.; Shneider, M. M.; Kostyuchenko, V. A.;

Mesyanzhinov, V. V.; Rossmann, M. G. The Tail Sheath Structure of Bacteriophage T4: A Molecular

Machine for Infecting Bacteria. EMBO J., 2009, 28 (7), 821–829.

https://doi.org/10.1038/emboj.2009.36.

[62] Leiman, P. G.; Chipman, P. R.; Kostyuchenko, V. A.; Mesyanzhinov, V. V.; Rossmann, M. G. Three-

Dimensional Rearrangement of Proteins in the Tail of Bacteriophage T4 on Infection of Its Host.

Cell, 2004, 118 (4), 419–429. https://doi.org/10.1016/j.cell.2004.07.022.

[63] Olszak, T.; Latka, A.; Roszniowski, B.; Valvano, M. A.; Drulis-Kawa, Z. Phage Life Cycles Behind

Bacterial Biodiversity. Curr. Med. Chem., 2017, 24 (36), 3987–4001.

https://doi.org/10.2174/0929867324666170413100136.

[64] Rakhuba, D. V.; Kolomiets, E. I.; Dey, E. S.; Novik, G. I. Bacteriophage Receptors, Mechanisms of

Phage Adsorption and Penetration into Host Cell. Polish J. Microbiol., 2010, 59 (3), 145–155.

https://doi.org/10.33073/pjm-2010-023.

[65] De Smet, J.; Hendrix, H.; Blasdel, B. G.; Danis-Wlodarczyk, K.; Lavigne, R. Pseudomonas

Predators: Understanding and Exploiting Phage-Host Interactions. Nat. Rev. Microbiol., 2017, 15

(9), 517–530. https://doi.org/10.1038/nrmicro.2017.61.

[66] Rakonjac, J.; Bennett, N. J.; Spagnuolo, J.; Gagic, D.; Russel, M. Filamentous Bacteriophage:

Biology, Phage Display and Nanotechnology Applications. Curr. Issues Mol. Biol., 2011, 13 (2), 51–

76. https://doi.org/10.21775/cimb.013.051.

[67] Hay, I. D.; Lithgow, T. Filamentous Phages: Masters of a Microbial Sharing Economy. EMBO Rep.,

2019, 20 (6), e47427. https://doi.org/10.15252/embr.201847427.

[68] Warwick-Dugdale, J.; Buchholz, H. H.; Allen, M. J.; Temperton, B. Host-Hijacking and Planktonic

Piracy: How Phages Command the Microbial High Seas. Virol. J., 2019, 16 (1), 15.

https://doi.org/10.1186/s12985-019-1120-1.

[69] Kutter, E.; Bryan, D.; Ray, G.; Brewster, E.; Blasdel, B.; Guttman, B. From Host to Phage

Metabolism: Hot Tales of Phage T4’s Takeover of E. Coli. Viruses, 2018, 10 (7), 387.

https://doi.org/10.3390/v10070387.

[70] Young, R. Phage Lysis: Three Steps, Three Choices, One Outcome. J. Microbiol., 2014, 52 (3), 243–

258. https://doi.org/10.1007/s12275-014-4087-z.


43

[71] Howard-Varona, C.; Hargreaves, K. R.; Abedon, S. T.; Sullivan, M. B. Lysogeny in Nature:

Mechanisms, Impact and Ecology of Temperate Phages. ISME J., 2017, 11 (7), 1511–1520.

https://doi.org/10.1038/ismej.2017.16.

[72] Maurice, C. F.; Bouvier, C.; De Wit, R.; Bouvier, T. Linking the Lytic and Lysogenic Bacteriophage

Cycles to Environmental Conditions, Host Physiology and Their Variability in Coastal Lagoons.

Environ. Microbiol., 2013, 15 (9), 2463–2475. https://doi.org/10.1111/1462-2920.12120.

[73] Coleman, S.; Yao, T.; Nguyen, T. V. P.; Golding, I.; Igoshin, O. Bacteriophage Self -Counting in the

Presence of Viral Replication. bioRxiv, 2021.

[74] Knowles, B.; Silveira, C. B.; Bailey, B. A.; Barott, K.; Cantu, V. A.; Cobia n-Guëmes, A. G.;

Coutinho, F. H.; Dinsdale, E. A.; Felts, B.; Furby, K. A.; et al. Lytic to Temperate Switching of Viral

Communities. Nature, 2016, 531 (7595), 466–470. https://doi.org/10.1038/nature17193.

[75] Payet, J. P.; Suttle, C. A. To Kill or Not to Kill: The Balance between Lytic and Lysogenic Viral

Infection Is Driven by Trophic Status. Limnol. Oceanogr., 2013, 58 (2), 465–474.

https://doi.org/10.4319/lo.2013.58.2.0465.

[76] Erez, Z.; Steinberger-Levy, I.; Shamir, M.; Doron, S.; Stokar-Avihail, A.; Peleg, Y.; Melamed, S.;

Leavitt, A.; Savidor, A.; Albeck, S.; et al. Communication between Viruses Guides Lysis–Lysogeny

Decisions. Nature, 2017, 541 (7638), 488–493. https://doi.org/10.1038/nature21049.

[77] Campbell, A. The Future of Bacteriophage Biology. Nat. Rev. Genet., 2003, 4 (6), 471–477.

https://doi.org/10.1038/nrg1089.

[78] Casjens, S. R.; Hendrix, R. W. Bacteriophage Lambda: Early Pioneer and Still Relevant. Virology,

2015, 479–480, 310–330. https://doi.org/10.1016/j.virol.2015.02.010.

[79] Zhang, Y.; Liao, Y. Te; Salvador, A.; Sun, X.; Wu, V. C. H. Prediction, Diversity, and Genomic

Analysis of Temperate Phages Induced From Shiga Toxin-Producing Escherichia Coli Strains.


[80] Nanda, A. M.; Thormann, K.; Frunzke, J. Impact of Spontaneous Prophage Induction on the Fitness

of Bacterial Populations and Host-Microbe Interactions. J. Bacteriol., 2015, 197 (3), 410–419.

https://doi.org/10.1128/JB.02230-14.

[81] Czyz, A.; Los, M.; Wrobel, B.; Wegrzyn, G. Inhibition of Spontaneous Induction of Lambdoid

Prophages in Escherichia Coli Cultures: Simple Procedures with Possible Biotechnological

Applications. BMC Biotechnol., 2001, 1, 1. https://doi.org/10.1186/1472-6750-1-1.

[82] Brussow, H.; Canchaya, C.; Hardt, W.-D. Phages and the Evolution of Bacterial Pathogens: From

Genomic Rearrangements to Lysogenic Conversion. Microbiol. Mol. Biol. Rev., 2004, 68 (3), 560–

602. https://doi.org/10.1128/MMBR.68.3.560-602.2004.

[83] Fortier, L.; Sekulovic, O. Importance of Prophages to Evolution and Virulence of Bacterial

Pathogens. Virulence, 2013, 4 (5), 354–365. https://doi.org/10.4161/viru.24498.

[84] Łoś, M.; Węgrzyn, G. Pseudolysogeny. In Advances in Virus Research; 2012; Vol. 82, pp 339–349.

https://doi.org/10.1016/B978-0-12-394621-8.00019-4.

[85] Cenens, W.; Makumi, A.; Mebrhatu, M. T.; Lavigne, R.; Aertsen, A. Phage–Host Interactions during

Pseudolysogeny. Bacteriophage, 2013, 3 (1), e25029. https://doi.org/10.4161/bact.25029.

[86] Ackermann, H. W. Phage Classification and Characterization. In Bacteriophages: Methods and

Protocols, Volume 1: Isolation, Characterization, and Interactions; Clokie, M. R. J., Kropinski, A.

M., Eds.; Methods in Molecular Biology; Humana Press: Totowa, NJ, 2009; Vol. 501, pp 127 –140.

https://doi.org/10.1007/978-1-60327-164-6.

[87] Rohwer, F.; Edwards, R. The Phage Proteomic Tree: A Genome-Based Taxonomy for Phage. J.

Bacteriol., 2002, 184 (16), 4529–4535. https://doi.org/10.1128/JB.184.16.4529-4535.2002.

[88] Adriaenssens, E. M.; Rodney Brister, J. How to Name and Classify Your Phage: An Informal Guide.


44

Viruses, 2017, 9 (4), 1–9. https://doi.org/10.3390/v9040070.

[89] Lefkowitz, E. J.; Dempsey, D. M.; Hendrickson, R. C.; Orton, R. J.; Siddell, S. G.; Smith, D. B.

Virus Taxonomy: The Database of the International Committee on Taxonomy of Viruses (ICTV).

Nucleic Acids Res., 2018, 46 (D1), D708–D717. https://doi.org/10.1093/nar/gkx932.

[90] Meier-Kolthoff, J. P.; Göker, M. VICTOR: Genome-Based Phylogeny and Classification of

Prokaryotic Viruses. Bioinformatics, 2017, 33 (21), 3396–3404.

https://doi.org/10.1093/bioinformatics/btx440.

[91] Gorbalenya, A. E.; Krupovic, M.; Mushegian, A.; Kropinski, A. M.; Siddell, S. G.; Varsani, A.;

Adams, M. J.; Davison, A. J.; Dutilh, B. E.; Harrach, B.; et al. The New Scope of Virus Taxonomy:

Partitioning the Virosphere into 15 Hierarchical Ranks. Nat. Microbiol., 2020, 5 (5), 668–674.

https://doi.org/10.1038/s41564-020-0709-x.

[92] Barylski, J.; Enault, F.; Dutilh, B. E.; Schuller, M. B.; Edwards, R. A.; Gillis, A.; Klumpp, J.;

Knezevic, P.; Krupovic, M.; Kuhn, J. H.; et al. Analysis of Spounaviruses as a Case Study for the

Overdue Reclassification of Tailed Phages. Syst. Biol., 2019, 69 (1), 110–123.

https://doi.org/10.1093/sysbio/syz036.

[93] Casjens, S. R.; Thuman-Commike, P. A. Evolution of Mosaically Related Tailed Bacteriophage

Genomes Seen through the Lens of Phage P22 Virion Assembly. Virology, 2011, 411 (2), 393–415.


[94] Lopes, A.; Tavares, P.; Petit, M. A.; Guérois, R.; Zinn-Justin, S. Automated Classification of Tailed

Bacteriophages According to Their Neck Organization. BMC Genomics, 2014, 15, 1027.

https://doi.org/10.1186/1471-2164-15-1027.

[95] Wilhem, S. W.; Poorvin, L. Quantification of Algal Viruses in Marine Samples. In Methods in

Microbiology; Paul, J. H., Ed.; 2001; Vol. 30, pp 53–65. https://doi.org/10.1016/S0580-

9517(01)30039-9.

[96] Anderson, B.; Rashid, M. H.; Carter, C.; Pasternack, G.; Rajanna, C.; Revazishvili, T.; Dean, T.;

Senecal, A.; Sulakvelidze, A. Enumeration of Bacteriophage Particles. Bacteriophage, 2011, 1 (2),

86–93. https://doi.org/10.4161/bact.1.2.15456.

[97] Ács, N.; Gambino, M.; Brøndsted, L. Bacteriophage Enumeration and Detection Methods. Front.

Microbiol., 2020, 11, 594868. https://doi.org/10.3389/fmicb.2020.594868.

[98] Snyder, J. C.; Spuhler, J.; Wiedenheft, B.; Roberto, F. F.; Douglas, T.; Young, M. J. Effects of

Culturing on the Population Structure of a Hyperthermophilic Virus. Microb. Ecol., 2004, 48 (4),

561–566. https://doi.org/10.1007/s00248-004-0246-9.

[99] Hatfull, G. F. Mycobacteriophages: Windows into Tuberculosis. PLoS Pathog., 2014, 10 (3), 12–14.

https://doi.org/10.1371/journal.ppat.1003953.

[100] Olsen, N. S.; Hendriksen, N. B.; Hansen, L. H.; Kot, W. A New High-Throughput Screening Method

for Phages: Enabling Crude Isolation and Fast Identification of Diverse Phages with Therapeutic

Potential. Phage, 2020, 1 (3), 137–148. https://doi.org/10.1089/phage.2020.0016.

[101] Mokili, J. L.; Rohwer, F.; Dutilh, B. E. Metagenomics and Future Perspectives in Virus Discovery.

Curr. Opin. Virol., 2012, 2 (1), 63–77. https://doi.org/10.1016/j.coviro.2011.12.004.

[102] Bharti, R.; Grimm, D. G. Current Challenges and Best-Practice Protocols for Microbiome Analysis.

Brief. Bioinform., 2021, 22 (1), 178–193. https://doi.org/10.1093/bib/bbz155.

[103] De Maio, N.; Shaw, L. P.; Hubbard, A.; George, S.; Sanderson, N. D.; Swann, J.; Wick, R.; Oun, M.

A.; Stubberfield, E.; Hoosdally, S. J.; et al. Comparison of Long-Read Sequencing Technologies in

the Hybrid Assembly of Complex Bacterial Genomes. Microb. Genomics, 2019, 5 (9), e000294.

https://doi.org/10.1099/mgen.0.000294.

[104] Mantere, T.; Kersten, S.; Hoischen, A. Long-Read Sequencing Emerging in Medical Genetics. Front.

Genet., 2019, 10, 426. https://doi.org/10.3389/fgene.2019.00426.


45

[105] Tyler, A. D.; Mataseje, L.; Urfano, C. J.; Schmidt, L.; Antonation, K. S.; Mulvey, M. R.; Corbett, C.

R. Evaluation of Oxford Nanopore’s MinION Sequencing Device for Microbial Whole Genome

Sequencing Applications. Sci. Rep., 2018, 8, 10931. https://doi.org/10.1038/s41598-018-29334-5.

[106] Roumpeka, D. D.; Wallace, R. J.; Escalettes, F.; Fotheringham, I.; Watson, M. A Review of

Bioinformatics Tools for Bio-Prospecting from Metagenomic Sequence Data. Front. Genet., 2017,

8, 23. https://doi.org/10.3389/fgene.2017.00023.

[107] Pearman, W. S.; Freed, N. E.; Silander, O. K. Testing the Advantages and Disadvantages of Short-

And Long- Read Eukaryotic Metagenomics Using Simulated Reads. BMC Bioinformatics, 2020, 21

(1), 220. https://doi.org/10.1186/s12859-020-3528-4.

[108] Gawad, C.; Koh, W.; Quake, S. R. Single-Cell Genome Sequencing: Current State of the Science.

Nat. Rev. Genet., 2016, 17 (3), 175–188. https://doi.org/10.1038/nrg.2015.16.

[109] Paul, J. H.; Sullivan, M. B.; Segall, A. M.; Rohwer, F. Marine Phage Genomics. Comp. Biochem.

Physiol. - B Biochem. Mol. Biol., 2002, 133 (4), 463–476. https://doi.org/10.1016/S1096-

4959(02)00168-9.

[110] Sullivan, M. B.; Coleman, M. L.; Quinlivan, V.; Rosenkrantz, J. E.; DeFrancesco, A. S.; Ta n, G.; Fu,

R.; Lee, J. A.; Waterbury, J. B.; Bielawski, J. P.; et al. Portal Protein Diversity and Phage Ecology.


[111] Adriaenssens, E. M.; Cowan, D. A. Using Signature Genes as Tools To Assess Environmental Viral

Ecology and Diversity. Appl. Environ. Microbiol., 2014, 80 (15), 4470–4480.

https://doi.org/10.1128/AEM.00878-14.

[112] Ackermann, H. W. Tailed Bacteriophages: The Order Caudovirales. Adv. Virus Res., 1998, 51, 135–

201.

[113] Roy, A.; Bhardwaj, A.; Datta, P.; Lander, G. C.; Cingolani, G. Small Terminase Couples Viral DNA

Binding to Genome-Packaging ATPase Activity. Structure, 2012, 20 (8), 1403–1413.

https://doi.org/10.1016/j.str.2012.05.014.

[114] Zhong, Y.; Chen, F.; Wilhelm, S. W.; Poorvin, L.; Hodson, R. E. Phylogenetic Diversity of Marine

Cyanophage Isolates and Natural Virus Communities as Revealed by Sequences of Viral Capsid

Assembly Protein Gene G20. In Applied and Environmental Microbiology; 2001; Vol. 30, pp 1576–

1584. https://doi.org/10.1128/AEM.68.4.1576-1584.2002.

[115] Comeau, A. M.; Krisch, H. M. The Capsid of the T4 Phage Superfamily: The Evolution, Diversity,

and Structure of Some of the Most Prevalent Proteins in the Biosphere. Mol. Biol. Evol., 2008, 25

(7), 1321–1332. https://doi.org/10.1093/molbev/msn080.

[116] Filée, J.; Tétart, F.; Suttle, C. A.; Krisch, H. M. Marine T4-Type Bacteriophages, a Ubiquitous

Component of the Dark Matter of the Biosphere. Proc. Natl. Acad. Sci. U. S. A., 2005, 102 (35),

12471–12476. https://doi.org/10.1073/pnas.0503404102.

[117] Dwivedi, B.; Schmieder, R.; Goldsmith, D. B.; Edwards, R. A.; Breitbart, M. PhiSiGns: An Online

Tool to Identify Signature Genes in Phages and Design PCR Primers for Examining Phage Diversity.

BMC Bioinformatics, 2012, 13 (1), 37. https://doi.org/10.1186/1471-2105-13-37.

[118] Gołębiewski, M.; Tretyn, A. Generating Amplicon Reads for Microbial Community Assessment

with Next-Generation Sequencing. J. Appl. Microbiol., 2019, 128 (2), 330–354.

https://doi.org/10.1111/jam.14380.

[119] Hsieh, S. Y.; Tariq, M. A.; Telatin, A.; Ansorge, R.; Adriaenssens, E. M.; Savva, G. M.; Booth, C.;

Wileman, T.; Hoyles, L.; Carding, S. R. Comparison of PCR versus PCR-Free DNA Library

Preparation for Characterising the Human Faecal Virome. Viruses, 2021, 13 (10), 2093.

https://doi.org/10.3390/v13102093.

[120] Mavrich, T. N.; Hatfull, G. F. Bacteriophage Evolution Differs by Host, Lifestyle and Genome. Nat.

Microbiol., 2017, 2, 17112. https://doi.org/10.1038/nmicrobiol.2017.112.


46

[121] Glazko, G.; Makarenkov, V.; Liu, J.; Mushegian, A. Evolutionary History of Bacteriophages with

Double-Stranded DNA Genomes. Biol. Direct, 2007, 2 (36), 1–14. https://doi.org/10.1186/1745-

6150-2-36.

[122] De Paepe, M.; Hutinet, G.; Son, O.; Amarir-Bouhram, J.; Schbath, S.; Petit, M. A. Temperate Phages

Acquire DNA from Defective Prophages by Relaxed Homologous Recombination: The Role of

Rad52-Like Recombinases. PLoS Genet., 2014, 10 (3).

https://doi.org/10.1371/journal.pgen.1004181.

[123] Botstein, D. A Theory of Modular Evolution for Bacteriophages. Ann. N. Y. Acad. Sci., 1980, 354,

484–491. https://doi.org/10.1111/j.1749-6632.1980.tb27987.x.

[124] Labrie, S. J.; Moineau, S. Abortive Infection Mechanisms and Prophage Sequences Significantly

Influence the Genetic Makeup of Emerging Lytic Lactococcal Phages. J. Bacteriol., 2007, 189 (4),

1482–1487. https://doi.org/10.1128/JB.01111-06.

[125] Lawrence, J. G.; Hatfull, G. F.; Hendrix, R. W. Imbroglios of Viral Taxonomy: Genetic Exchange

and Failings of Phenetic Approaches. J. Bacteriol., 2002, 184 (17), 4891–4905.

https://doi.org/10.1128/JB.184.17.4891-4905.2002.

[126] Hendrix, R. W. Bacteriophages: Evolution of the Majority. Theor. Popul. Biol., 2002, 61 (4), 471–

480. https://doi.org/10.1006/tpbi.2002.1590.

[127] Pedulla, M. L.; Ford, M. E.; Houtz, J. M.; Karthikeyan, T.; Wadsworth, C.; Lewis, J. A.; Jacobs-

Sera, D.; Falbo, J.; Gross, J.; Pannunzio, N. R.; et al. Origins of Highly Mosaic Mycobacteriophage

Genomes. Cell, 2003, 113 (2), 171–182. https://doi.org/10.1016/S0092-8674(03)00233-2.

[128] Moura de Sousa, J. A.; Pfeifer, E.; Touchon, M.; Rocha, E. P. C. Causes and Consequences of

Bacteriophage Diversification via Genetic Exchanges across Lifestyles and Bacterial Taxa. Mol.

Biol. Evol., 2021, 38 (6), 2497–2512. https://doi.org/10.1093/molbev/msab044.

[129] Pope, W. H.; Bowman, C. A.; Russell, D. A.; Jacobs-Sera, D.; Asai, D. J.; Cresawn, S. G.; Jacobs,

W. R.; Hendrix, R. W.; Lawrence, J. G.; Hatfull, G. F. Whole Genome Comparison of a Large

Collection of Mycobacteriophages Reveals a Continuum of Phage Genetic Diversity. Elife, 2015, 4,

e06416. https://doi.org/10.7554/elife.06416.

[130] Viertel, T. M.; Ritter, K.; Horz, H. P. Viruses versus Bacteria -Novel Approaches to Phage Therapy

as a Tool against Multidrug-Resistant Pathogens. J. Antimicrob. Chemother., 2014, 69 (9), 2326–

2336. https://doi.org/10.1093/jac/dku173.

[131] Rohde, C.; Wittmann, J.; Kutter, E. Bacteriophages: A Therapy Concept against Multi-Drug-

Resistant Bacteria. Surg. Infect. (Larchmt)., 2018, 19 (8), 737–744.

https://doi.org/10.1089/sur.2018.184.

[132] Kortright, K. E.; Chan, B. K.; Koff, J. L.; Turner, P. E. Phage Therapy: A Renewed Approach to

Combat Antibiotic-Resistant Bacteria. Cell Host Microbe, 2019, 25 (2), 219–232.

https://doi.org/10.1016/j.chom.2019.01.014.

[133] Lu, T. K.; Koeris, M. S. The next Generation of Bacteriophage Therapy. Curr. Opin. Microbiol.,

2011, 14 (5), 524–531. https://doi.org/10.1016/j.mib.2011.07.028.

[134] Gibb, B.; Hyman, P.; Schneider, C. L. The Many Applications of Engineered Bacteriophages—An

Overview. Pharmaceuticals, 2021, 14 (7), 634. https://doi.org/10.3390/ph14070634.

[135] Hatfull, G. F.; Hendrix, R. W. Bacteriophages and Their Genomes. Curr. Opin. Virol., 2011, 1 (4),

298–303. https://doi.org/10.1016/j.coviro.2011.06.009.

[136] Hayes, S.; Mahony, J.; Nauta, A.; Van Sinderen, D. Metagenomic Approaches to Assess

Bacteriophages in Various Environmental Niches. Viruses, 2017, 9 (6), 127.

https://doi.org/10.3390/v9060127.

[137] Ofir, G.; Sorek, R. Contemporary Phage Biology: From Classic Models to New Insights. Cell, 2018,

172 (6), 1260–1270. https://doi.org/10.1016/j.cell.2017.10.045.


47

[138] Harada, L. K.; Silva, E. C.; Campos, W. F.; Del Fiol, F. S.; Vila, M.; Dąbrowska, K.; Krylov, V. N.;

Balcão, V. M. Biotechnological Applications of Bacteriophages: State of the Art. Microbiol. Res.,

2018, 212–213, 38–58. https://doi.org/10.1016/j.micres.2018.04.007.

[139] Moye, Z. D.; Woolston, J.; Sulakvelidze, A. Bacteriophage Applications for Food Production and

Processing. Viruses, 2018, 10 (4), 205. https://doi.org/10.3390/v10040205.

[140] Nikolich, M. P.; Filippov, A. A. Bacteriophage Therapy: Developments and Directions. Antibiotics,

2020, 9 (3). https://doi.org/10.3390/antibiotics9030135.

[141] Monk, A. B.; Rees, C. D.; Barrow, P.; Hagens, S.; Harper, D. R. Bacteriophage Applications: Where

Are We Now? Lett. Appl. Microbiol., 2010, 51 (4), 363–369. https://doi.org/10.1111/j.1472-

765X.2010.02916.x.

[142] Toribio-Avedillo, D.; Blanch, A. R.; Muniesa, M.; Rodríguez-Rubio, L. Bacteriophages as Fecal

Pollution Indicators. Viruses, 2021, 13 (6), 1089. https://doi.org/10.3390/v13061089.

[143] Fair, R. J.; Tor, Y. Perspectives in Medicinal Chemistry Antibiotics and Bacterial Resistance in the

21st Century. Perspect. Medicin. Chem., 2014, 25–64.

https://doi.org/10.4137/PMC.S14459.Received.

[144] Loc-Carrillo, C.; Abedon, S. T. Pros and Cons of Phage Therapy. Bacteriophage, 2011, 1 (2), 111–

114. https://doi.org/10.4161/bact.1.2.14590.

[145] Principi, N.; Silvestri, E.; Esposito, S. Advantages and Limitations of Bacteriophages for the

Treatment of Bacterial Infections. Front. Pharmacol., 2019, 10, 513.

https://doi.org/10.3389/fphar.2019.00513.

[146] Sulakvelidze, A.; Barrow, P. Phage Therapy in Animals and Agribusiness. In Bacteriophages:

Biology and Applications; Kutter, E., Sulakvelidze, A., Eds.; CRC Press: Boca Raton, FL, 2005; pp

335–380.

[147] Brüssow, H. Phage Therapy: The Escherichia Coli Experience. Microbiology, 2005, 151 (7), 2133–

2140. https://doi.org/10.1099/mic.0.27849-0.

[148] Ferriol-González, C.; Domingo-Calap, P. Phages for Biofilm Removal. Antibiotics, 2020, 9 (5), 268.

https://doi.org/10.3390/antibiotics9050268.

[149] Harper, D. R.; Parracho, H. M. R. T.; Walker, J.; Sharp, R.; Hughes, G.; Werthén, M.; Lehman, S.;

Morales, S. Bacteriophages and Biofilms. Antibiotics, 2014, 3 (3), 270–284.


[150] Sharma, U.; Vipra, A.; Channabasappa, S. Phage-Derived Lysins as Potential Agents for Eradicating

Biofilms and Persisters. Drug Discov. Today, 2018, 23 (4), 848–856.

https://doi.org/10.1016/j.drudis.2018.01.026.

[151] Hyman, P.; Abedon, S. T. Bacteriophage Host Range and Bacterial Resistance. In Advances in

Applied Microbiolog; Elsevier Inc., 2010; Vol. 70, pp 217–248. https://doi.org/10.1016/S0065-

2164(10)70007-1.

[152] Kutateladze, M.; Adamia, R. Bacteriophages as Potential New Therapeutics to Replace or

Supplement Antibiotics. Trends Biotechnol., 2010, 28 (12), 591–595.

https://doi.org/10.1016/j.tibtech.2010.08.001.

[153] Wright, R. C. T.; Friman, V. P.; Smith, M. C. M.; Brockhurst, M. A. Resistance Evolution against

Phage Combinations Depends on the Timing and Order of Exposure. MBio, 2019, 10 (5), e01652-

19. https://doi.org/10.1128/mBio.01652-19.

[154] Buckling, A.; Rainey, P. B. Antagonistic Coevolution between a Bacterium and a Bacteriophage.

Proc. R. Soc. B Biol. Sci., 2002, 269 (1494), 931–936. https://doi.org/10.1098/rspb.2001.1945.

[155] Laanto, E.; Hoikkala, V.; Ravantti, J.; Sundberg, L.-R. Long-Term Genomic Coevolution of Host-

Parasite Interaction in the Natural Environment. Nat. Commun., 2017, 8, 111.


48

https://doi.org/10.1038/s41467-017-00158-7.

[156] León, M.; Bastías, R. Virulence Reduction in Bacteriophage Resistant Bacteria. Front. Microbiol.,

2015, 6 (343), 1–7. https://doi.org/10.3389/fmicb.2015.00343.

[157] Chan, B. K.; Abedon, S. T. Phage Therapy Pharmacology: Phage Cocktails. In Advances in Applied

Microbiology; Elsevier Inc., 2012; Vol. 78, pp 1–23. https://doi.org/10.1016/B978-0-12-394805-

2.00001-4.

[158] Bull, J. J.; Gill, J. J. The Habits of Highly Effective Phages: Population Dynamics as a Framework

for Identifying Therapeutic Phages. Front. Microbiol., 2014, 5 (618).


[159] Skurnik, M.; Pajunen, M.; Kiljunen, S. Biotechnological Challenges of Phage Therapy. Biotechnol.

Lett., 2007, 29 (7), 995–1003. https://doi.org/10.1007/s10529-007-9346-1.

[160] Gill, J.; Hyman, P. Phage Choice, Isolation, and Preparation for Phage Therapy. Curr. Pharm.

Biotechnol., 2010, 11 (1), 2–14. https://doi.org/10.2174/138920110790725311.

[161] Bretaudeau, L.; Tremblais, K.; Aubrit, F.; Meichenin, M.; Arnaud, I. Good Manufacturing Practice

(GMP) Compliance for Phage Therapy Medicinal Products. Front. Microbiol., 2020, 11, 1161.


[162] Dedrick, R. M.; Jacobs-Sera, D.; Bustamante, C. A. G.; Garlena, R. A.; Mavrich, T. N.; Pope, W.

H.; Reyes, J. C. C.; Russell, D. A.; Adair, T.; Alvey, R.; et al. Prophage-Mediated Defence against

Viral Attack and Viral Counter-Defence. Nat. Microbiol., 2017, 2, 16251.

https://doi.org/10.1038/nmicrobiol.2016.251.

[163] Monteiro, R.; Pires, D. P.; Costa, A. R.; Azeredo, J. Phage Therapy: Going Temperate? Trends

Microbiol., 2019, 27 (4), 368–378. https://doi.org/10.1016/j.tim.2018.10.008.

[164] Goodridge, L. Designing Phage Therapeutics. Curr. Pharm. Biotechnol., 2010, 11 (1), 15–27.

https://doi.org/10.2174/138920110790725348.

[165] Lu, T. K.; Collins, J. J. Dispersing Biofilms with Engineered Enzymatic Bacteriophage. Proc. Natl.

Acad. Sci. U. S. A., 2007, 104 (27), 11197–11202. https://doi.org/10.1073/pnas.0704624104.

[166] Ando, H.; Lemire, S.; Pires, D. P.; Lu, T. K. Engineering Modular Viral Scaffolds for Targeted

Bacterial Population Editing. Cell Syst., 2015, 1 (3), 187–196.

https://doi.org/10.1016/j.cels.2015.08.013.

[167] Lemire, S.; Yehl, K. M.; Lu, T. K. Phage-Based Applications in Synthetic Biology. Annu. Rev. Virol.,

2018, 5 (1), 453–476. https://doi.org/10.1146/annurev-virology-092917-043544.

[168] Hargreaves, K. R.; Clokie, M. R. J. Clostridium Difficile Phages: Still Difficult? Front. Microbiol.,

2014, 5, 184. https://doi.org/10.3389/fmicb.2014.00184.

[169] Meile, S.; Du, J.; Dunne, M.; Kilcher, S.; Loessner, M. J. Engineering Therapeutic Phages for

Enhanced Antibacterial Efficacy. Curr. Opin. Virol., 2022, 52, 182–191.

https://doi.org/10.1016/j.coviro.2021.12.003.

[170] Pires, D. P.; Cleto, S.; Sillankorva, S.; Azeredo, J.; Lu, T. K. Genetically Engineered Phages: A

Review of Advances over the Last Decade. Microbiol. Mol. Biol. Rev., 2016, 80 (3), 523–543.


[171] Sulakvelidze, A.; Alavidze, Z.; Morris, J. G. Bacteriophage Therapy. Antimicrob. Agents

Chemother., 2001, 45 (3), 649–659. https://doi.org/10.1128/AAC.45.3.649-659.2001.

[172] Roach, D. R.; Leung, C. Y.; Henry, M.; Morello, E.; Singh, D.; Di Santo, J. P.; Weitz, J. S.;

Debarbieux, L. Synergy between the Host Immune System and Bacteriophage Is Essential for

Successful Phage Therapy against an Acute Respiratory Pathogen. Cell Host Microbe, 2017, 22 (1),

38-47.e4. https://doi.org/10.1016/j.chom.2017.06.018.


49

[173] Van Belleghem, J.; Dąbrowska, K.; Vaneechoutte, M.; Barr, J.; Bollyky, P. Interactions between

Bacteriophage, Bacteria, and the Mammalian Immune System. Viruses, 2018, 11 (1), 10.

https://doi.org/10.3390/v11010010.

[174] Abedon, S. T.; Kuhl, S. J.; Blasdel, B. G.; Kutter, E. M. Phage Treatment of Human Infections.

Bacteriophage, 2011, 1 (2), 66–85. https://doi.org/10.4161/bact.1.2.15845.

[175] Labrie, S. J.; Samson, J. E.; Moineau, S. Bacteriophage Resistance Mechanisms. Nat. Rev.

Microbiol., 2010, 8 (5), 317–327. https://doi.org/10.1038/nrmicro2315.

[176] Oechslin, F. Resistance Development to Bacteriophages Occurring during Bacteriophage Therapy.

Viruses, 2018, 10 (7). https://doi.org/10.3390/v10070351.

[177] Taati Moghadam, M.; Amirmozafari, N.; Shariati, A.; Hallajzadeh, M.; Mirkalantari, S.;

Khoshbayan, A.; Masjedian Jazi, F. How Phages Overcome the Challenges of Drug Resistant

Bacteria in Clinical Infections. Infect. Drug Resist., 2020, 13, 45–61.

https://doi.org/10.2147/IDR.S234353.

[178] Nobrega, F. L.; Costa, A. R.; Kluskens, L. D.; Azeredo, J. Revisiting Phage Therapy: New

Applications for Old Resources. Trends Microbiol., 2015, 23 (4), 185–191.


[179] Tanji, Y.; Shimada, T.; Yoichi, M.; Miyanaga, K.; Hori, K.; Unno, H. Toward Rational Control of

Escherichia Coli O157:H7 by a Phage Cocktail. Appl. Microbiol. Biotechnol., 2004, 64 (2), 270–

274. https://doi.org/10.1007/s00253-003-1438-9.

[180] Örmälä, A.-M.; Jalasvuori, M. Phage Therapy: Should Bacterial Resistance to Phages Be a Concern,

Even in the Long Run? Bacteriophage, 2013, 3 (1), e24219. https://doi.org/10.4161/bact.24219.

[181] Chan, B. K.; Abedon, S. T.; Loc-Carrillo, C. Phage Cocktails and the Future of Phage Therapy.

Future Microbiol., 2013, 8 (6), 769–783. https://doi.org/10.2217/fmb.13.47.

[182] McCallin, S.; Sacher, J. C.; Zheng, J.; Chan, B. K. Current State of Compassionate Phage Therapy.

Viruses, 2019, 11 (4), 343. https://doi.org/10.3390/v11040343.

[183] Yerushalmy, O.; Khalifa, L.; Gold, N.; Rakov, C.; Alkalay-Oren, S.; Adler, K.; Ben-Porat, S.;

Kraitman, R.; Gronovich, N.; Ginat, K. S.; et al. The Israeli Phage Bank (IPB). Antibiotics, 2020, 9

(5), 269. https://doi.org/10.3390/antibiotics9050269.

[184] Tsonos, J.; Oosterik, L. H.; Tuntufye, H. N.; Klumpp, J.; Butaye, P.; De Greve, H.; Hernalsteens, J.

P.; Lavigne, R.; Goddeeris, B. M. A Cocktail of in Vitro Efficient Phages Is Not a Guarantee for in

Vivo Therapeutic Results against Avian Colibacillosis. Vet. Microbiol., 2014.

https://doi.org/10.1016/j.vetmic.2013.10.021.

[185] Hyman, P.; Abedon, S. T. Practical Methods for Determining Phage Growth Parameters. In

Bacteriophages. Methods in Molecular Biology; Clokie, M. R. J., Kropinski, A. M., Eds.; Methods

in Molecular Biology; Humana Press: Totowa, NJ, 2009; Vol. 501, pp 175 –202.

https://doi.org/10.1007/978-1-60327-164-6.

[186] T. Abedon, S. Phage “Delay” towards Enhancing Bacterial Escape from Biofilms: A More

Comprehensive Way of Viewing Resistance to Ba cteriophages. AIMS Microbiol., 2017, 3 (2), 186–

226. https://doi.org/10.3934/microbiol.2017.2.186.

[187] Casey, E.; van Sinderen, D.; Mahony, J. In Vitro Characteristics of Phages to Guide ‘Real Life’

Phage Therapy Suitability. Viruses, 2018, 10 (4), 163. https://doi.org/10.3390/v10040163.

[188] Levin, B. R.; Bull, J. J. Population and Evolutionary Dynamics of Phage Therapy. Nat. Rev.


[189] Pirnay, J. P.; Verbeken, G.; Ceyssens, P. J.; Huys, I.; de Vos, D.; Ameloot, C.; Fauconnier, A. The

Magistral Phage. Viruses, 2018, 10 (2), 64. https://doi.org/10.3390/v10020064.

[190] Mangalea, M. R.; Duerkop, B. A. Fitness Trade-Offs Resulting from Bacteriophage Resistance


50

Potentiate Synergistic Antibacterial Strategies. Infect. Immun., 2020, 88 (7), e00926-19.

https://doi.org/10.1128/IAI.00926-19.

[191] Huff, W. E.; Huff, G. R.; Rath, N. C.; Balog, J. M.; Donoghue, A. M. Therapeutic Efficacy of

Bacteriophage and Baytril (Enrofloxacin) Individually and in Combination to Treat Colibacillosis in

Broilers. Poult. Sci., 2004, 83 (12), 1944–1947. https://doi.org/10.1093/ps/83.12.1944.

[192] Lu, T. K.; Collins, J. J. Engineered Bacteriophage Targeting Gene Networks as Adjuvants for

Antibiotic Therapy. Proc. Natl. Acad. Sci., 2009, 106 (12), 4629–4634.

https://doi.org/10.1073/pnas.0800442106.

[193] Koskella, B.; Brockhurst, M. A. Bacteria –Phage Coevolution as a Driver of Ecological and

Evolutionary Processes in Microbial Communities. FEMS Microbiol. Rev., 2014, 38 (5), 916–931.

https://doi.org/10.1111/1574-6976.12072.

[194] Secor, P. R.; Dandekar, A. A. More than Simple Parasites: The Sociobiology of Bacteriophages and

Their Bacterial Hosts. MBio, 2020, 11 (2), 1–13. https://doi.org/10.1128/mBio.00041-20.

[195] Weinbauer, M. G. Ecology of Prokaryotic Viruses. FEMS Microbiol. Rev., 2004, 28 (2), 127–181.

https://doi.org/10.1016/j.femsre.2003.08.001.

[196] Chaturongakul, S.; Ounjai, P. Phage-Host Interplay: Examples from Tailed Phages and Gram -

Negative Bacterial Pathogens. Front. Microbiol., 2014, 5, 442.


[197] Sausset, R.; Petit, M. A.; Gaboriau-Routhiau, V.; De Paepe, M. New Insights into Intestinal Phages.

Mucosal Immunol., 2020, 13 (2), 205–215. https://doi.org/10.1038/s41385-019-0250-5.

[198] Obeng, N.; Pratama, A. A.; Elsas, J. D. van. The Significance of Mutualistic Phages for Bacterial

Ecology and Evolution. Trends Microbiol., 2016, 24 (6), 440–449.


[199] Labrie, S. J.; Moineau, S. Phosphorylation, an Altruistic Bacterial Trick to Halt Phages. Cell Host

Microbe, 2016, 20 (4), 409–410. https://doi.org/10.1016/j.chom.2016.09.017.

[200] Chibani-chennoufi, S.; Bruttin, A.; Brüssow, H.; Dillmann, M.; Bru, H. Phage-Host Interaction : An

Ecological Perspective. J. Bacteriol., 2004, 186 (12), 3677–3686.

https://doi.org/10.1128/JB.186.12.3677.

[201] Juliano, S. A. Population Dynamics. J. Am. Mosq. Control Assoc., 2007, 23 (2 Suppl), 265–275.

https://doi.org/10.2987/8756-971X(2007)23[265:PD]2.0.CO;2.

[202] Santos, S. B.; Carvalho, C.; Azeredo, J.; Ferreira, E. C. Population Dynamics of a Salmonella Lytic

Phage and Its Host: Implications of the Host Bacterial Growth Rate in Modelling. PLoS One, 2014,

9 (7), e102507. https://doi.org/10.1371/journal.pone.0102507.

[203] Krysiak-Baltyn, K.; Martin, G. J. O.; Stickland, A. D.; Scales, P. J.; Gras, S. L. Computational

Models of Populations of Bacteria and Lytic Phage. Crit. Rev. Microbiol., 2016, 42 (6), 942–968.

https://doi.org/10.3109/1040841X.2015.1114466.

[204] Bohannan, B. J. M.; Lenski, R. D. Linking Genetic Change to Community Evolution: Insights from

Studies of Bacteria and Bacteriophage. Ecol. Lett., 2000, 3 (4), 362–377.

https://doi.org/https://doi.org/10.1046/j.1461-0248.2000.00161.x.

[205] Weissman, J. L.; Holmes, R.; Barrangou, R.; Moineau, S.; Fagan, W. F.; Levin, B.; Johnson, P. L.

F. Immune Loss as a Driver of Coexistence during Host-Phage Coevolution. ISME J., 2018, 12 (2),

585–597. https://doi.org/10.1038/ismej.2017.194.

[206] Roucourt, B.; Lavigne, R. The Role of Interactions between Phage and Bacterial Proteins within the

Infected Cell: A Diverse and Puzzling Interactome. Environ. Microbiol., 2009, 11 (11), 2789–2805.

https://doi.org/10.1111/j.1462-2920.2009.02029.x.

[207] Mahmoudabadi, G.; Milo, R.; Phillips, R. Energetic Cost of Building a Virus. Proc. Natl. Acad. Sci.


51

U. S. A., 2017, 114 (22), E4324–E4333. https://doi.org/10.1073/pnas.1701670114.

[208] Stern, A.; Sorek, R. The Phage-Host Arms Race: Shaping the Evolution of Microbes. BioEssays,

2011, 33 (1), 43–51. https://doi.org/10.1002/bies.201000071.

[209] Holguín, A.; Cárdenas, P.; Prada -Peñaranda, C.; Rabelo Leite, L.; Buitrago, C.; Clavijo, V.; Oliveira,

G.; Leekitcharoenphon, P.; Møller Aarestrup, F.; Vives, M. Host Resistance, Genomics and

Population Dynamics in a Salmonella Enteritidis and Phage System . Viruses, 2019, 11 (2), 188.

https://doi.org/10.3390/v11020188.

[210] Gómez, P.; Buckling, A. Bacteria -Phage Antagonistic Coevolution in Soil. Science (80-. )., 2011,

332 (6025), 106–109. https://doi.org/10.1126/science.1198767.

[211] Gorter, F. A.; Scanlan, P. D.; Buckling, A. Adaptation to Abiotic Conditions Drives Local Adaptation

in Bacteria and Viruses Coevolving in Heterogeneous Environments. Biol. Lett., 2016, 12, 20150879.

https://doi.org/10.1098/rsbl.2015.0879.

[212] Storms, Z. J.; Sauvageau, D. Modeling Tailed Bacteriophage Adsorption: Insight into Mechanisms.

Virology, 2015, 485, 355–362. https://doi.org/10.1016/j.virol.2015.08.007.

[213] Storms, Z. J.; Teel, M. R.; Mercurio, K.; Sauvageau, D. The Virulence Index: A Metric for

Quantitative Analysis of Phage Virulence. Phage, 2020, 1 (1), 27–36.

https://doi.org/10.1089/phage.2019.0001.

[214] Cairns, B. J.; Timms, A. R.; Jansen, V. A. A.; Connerton, I. F.; Payne, R. J. H. Quantitative Models

of in Vitro Bacteriophage-Host Dynamics and Their Application to Phage Therapy. PLoS Pathog.,

2009, 5 (1), e1000253. https://doi.org/10.1371/journal.ppat.1000253.

[215] Yehl, K.; Lemire, S.; Yang, A. C.; Ando, H.; Mimee, M.; Torres, M. D. T.; de la Fuente-Nunez, C.;

Lu, T. K. Engineering Phage Host-Range and Suppressing Bacterial Resistance through Phage Tail

Fiber Mutagenesis. Cell, 2019, 179 (2), 459-469.e9. https://doi.org/10.1016/j.cell.2019.09.015.

[216] Gabig, M.; Herman-Antosiewicz, A.; Kwiatkowska, M.; Los, M.; Thomas, M. S.; Wȩgrzyn, G. The

Cell Surface Protein Ag43 Facilitates Phage Infection of Escherichia Coli in the Presence of Bile

Salts and Carborhydrates. Microbiology, 2002, 148 (5), 1533–1542.

https://doi.org/10.1099/00221287-148-5-1533.

[217] Park, S. H.; Biswas, D.; Lingbeck, J.; Koo, O. K.; Ricke, S. C. Enhancement of Chicken Macrophage

Cytokine Response to Salmonella Typhimurium When Combined with Bacteriophage P22. FEMS

Microbiol. Lett., 2013, 339 (2), 137–144. https://doi.org/10.1111/1574-6968.12064.

[218] Carroll-Portillo, A.; Lin, H. C. Bacteriophage and the Innate Immune System: Access and Signaling.

Microorganisms, 2019, 7 (12), 625. https://doi.org/10.3390/microorganisms7120625.

[219] Stone, E.; Campbell, K.; Grant, I.; McAuliffe, O. Understanding and Exploiting Phage–Host

Interactions. Viruses, 2019, 11 (6), 567. https://doi.org/10.3390/v11060567.

[220] Vale, P. F.; Little, T. J. CRISPR-Mediated Phage Resistance and the Ghost of Coevolution Past.

Proc. R. Soc. Biol. Sci., 2010, 277 (1691), 2097–2103. https://doi.org/10.1098/rspb.2010.0055.

[221] Goldfarb, T.; Sberro, H.; Weinstock, E.; Cohen, O.; Doron, S.; Charpak‐Amikam, Y.; Afik, S.; Ofir,

G.; Sorek, R. <scp>BREX</Scp> Is a Novel Phage Resistance System Widespread in Microbial

Genomes. EMBO J., 2015, 34 (2), 169–183. https://doi.org/10.15252/embj.201489455.

[222] Ofir, G.; Melamed, S.; Sberro, H.; Mukamel, Z.; Silverman, S.; Yaakov, G.; Doron, S.; Sorek, R.

DISARM Is a Widespread Bacterial Defence System with Broad Anti-Phage Activities. Nat.

Microbiol., 2018, 3 (1), 90–98. https://doi.org/10.1038/s41564-017-0051-0.

[223] Swanson, N. A.; Cingolani, G. A Tail of Phage Adhesins. Structure, 2018, 26 (12), 1565–1567.

https://doi.org/10.1016/j.str.2018.11.008.

[224] Letarov, A. V.; Kulikov, E. E. Adsorption of Bacteriophages on Bacterial Cells. Biochem., 2017, 82

(13), 1632–1658. https://doi.org/10.1134/S0006297917130053.


52

[225] Bertozzi Silva, J.; Storms, Z.; Sauvageau, D. Host Receptors for Bacteriophage Adsorption. FEMS

Microbiol. Lett., 2016, 363 (4), 1–11. https://doi.org/10.1093/femsle/fnw002.

[226] Islam, M. Z.; Fokine, A.; Mahalingam, M.; Zhang, Z.; Garcia -Doval, C.; Van Raaij, M. J.; Rossmann,

M. G.; Rao, V. B. Molecular Anatomy of the Receptor Binding Module of a Bacteriophage Long

Tail Fiber. PLoS Pathog., 2019, 15 (12), e1008193. https://doi.org/10.1371/journal.ppat.1008193.

[227] Maffei, E.; Shaidullina, A.; Burkolter, M.; Heyer, Y.; Estermann, F.; Druelle, V.; Sauer, P.; Willi,

L.; Michaelis, S.; Hilbi, H.; et al. Systematic Exploration of Escherichia Coli Phage–Host

Interactions with the BASEL Phage Collection. PLOS Biol., 2021, 19 (11), e3001424.

https://doi.org/10.1371/journal.pbio.3001424.

[228] Wu, L. T.; Chang, S. Y.; Yen, M. R.; Yang, T. C.; Tseng, Y. H. Characterization of Extended -Host-

Range Pseudo-T-Even Bacteriophage Kpp95 Isolated on Klebsiella Pneumoniae. Appl. Environ.

Microbiol., 2007, 73 (8), 2532–2540. https://doi.org/10.1128/AEM.02113-06.

[229] Nobrega, F. L.; Vlot, M.; de Jonge, P. A.; Dreesens, L. L.; Beaumont, H. J. E.; Lavigne, R.; Dutilh,

B. E.; Brouns, S. J. J. Targeting Mechanisms of Tailed Bacteriophages. Nat. Rev. Microbiol., 2018,

16 (12), 760–773. https://doi.org/10.1038/s41579-018-0070-8.

[230] Dimitriu, T.; Szczelkun, M. D.; Westra, E. R. Evolutionary Ecology and Interplay of Prokaryotic

Innate and Adaptive Immune Systems. Curr. Biol., 2020, 30 (19), R1189–R1202.

https://doi.org/10.1016/j.cub.2020.08.028.

[231] Vasu, K.; Nagaraja, V. Diverse Functions of Restriction-Modification Systems in Addition to

Cellular Defense. Microbiol. Mol. Biol. Rev., 2013, 77 (1), 53–72.


[232] Samson, J. E.; Magadán, A. H.; Sabri, M.; Moineau, S. Revenge of the Phages: Defeating Bacterial

Defences. Nat. Rev. Microbiol., 2013, 11 (10), 675–687. https://doi.org/10.1038/nrmicro3096.

[233] Nechaev, S.; Severinov, K. The Elusive Object of Desire—Interactions of Bacteriophages and Their

Hosts. Curr. Opin. Microbiol., 2008, 11 (2), 186–193. https://doi.org/10.1016/j.mib.2008.02.009.

[234] Lopatina, A.; Tal, N.; Sorek, R. Abortive Infection: Bacterial Suicide as an Antiviral Immune

Strategy. Annu. Rev. Virol., 2020, 7 (1), 371–384. https://doi.org/10.1146/annurev-virology-011620-

040628.

[235] Fineran, P. C.; Blower, T. R.; Foulds, I. J.; Humphreys, D. P.; Lilley, K. S.; Salmond, G. P. C. The

Phage Abortive Infection System, ToxIN, Functions as a Protein–RNA Toxin–Antitoxin Pair. Proc.

Natl. Acad. Sci., 2009, 106 (3), 894–899. https://doi.org/10.1073/pnas.0808832106.

[236] Gerdes, K.; Christensen, S. K.; Løbner-Olesen, A. Prokaryotic Toxin-Antitoxin Stress Response

Loci. Nat. Rev. Microbiol., 2005, 3 (5), 371–382. https://doi.org/10.1038/nrmicro1147.

[237] Chen, B.; Akusobi, C.; Fang, X.; Salmond, G. P. C. Environmental T4-Family Bacteriophages

Evolve to Escape Abortive Infection via Multiple Routes in a Bacterial Host Employing “Altruistic

Suicide” through Type III Toxin-Antitoxin Systems. Front. Microbiol., 2017, 8 (MAY), 1–11.


[238] Blower, T. R.; Chai, R.; Przybilski, R.; Chindhy, S.; Fang, X.; Kidman, S. E.; Tan, H.; Luisi, B. F.;

Fineran, P. C.; Salmond, G. P. C. Evolution of Pectobacterium Bacteriophage ΦM1 To Escape Two

Bifunctional Type III Toxin-Antitoxin and Abortive Infection Systems through Mutations in a Single

Viral Gene. Appl. Environ. Microbiol., 2017, 83 (8), 1–17. https://doi.org/10.1128/AEM.03229-16.

[239] Seed, K. D. Battling Phages: How Bacteria Defend against Viral Attack. PLOS Pathog., 2015, 11

(6), e1004847. https://doi.org/10.1371/journal.ppat.1004847.

[240] Blower, T. R.; Evans, T. J.; Przybilski, R.; Fineran, P. C.; Salmond, G. P. C. Viral Evasion of a

Bacterial Suicide System by RNA–Based Molecular Mimicry Enables Infectious Altruism. PLoS

Genet., 2012, 8 (10), e1003023. https://doi.org/10.1371/journal.pgen.1003023.

[241] Blower, T. R.; Short, F. L.; Fineran, P. C.; Salmond, G. P. C. Viral Molecular Mimicry Circumvents


53

Abortive Infection and Suppresses Bacterial Suicide to Make Hosts Permissive for Replication.

Bacteriophage, 2012, 2 (4), e23830. https://doi.org/10.4161/bact.23830.

[242] Pourcel, C.; Touchon, M.; Villeriot, N.; Vernadet, J.-P.; Couvin, D.; Toffano-Nioche, C.; Vergnaud,

G. CRISPRCasdb a Successor of CRISPRdb Containing CRISPR Arrays and Cas Genes from

Complete Genome Sequences, and Tools to Download and Query Lists of Repeats and Spacers.

Nucleic Acids Res., 2020, 48 (D1), D535–D544. https://doi.org/10.1093/nar/gkz915.

[243] Barrangou, R.; Marraffini, L. A. CRISPR-Cas Systems: Prokaryotes Upgrade to Adaptive Immunity.

Mol. Cell, 2014, 54 (2), 234–244. https://doi.org/10.1016/j.molcel.2014.03.011.

[244] Sorek, R.; Lawrence, C. M.; Wiedenheft, B. CRISPR-Mediated Adaptive Immune Systems in

Bacteria and Archaea. Annu. Rev. Biochem., 2013, 82 (1), 237–266. https://doi.org/10.1146/annurev-

biochem-072911-172315.

[245] Jackson, S. A.; McKenzie, R. E.; Fagerlund, R. D.; Kieper, S. N.; Fineran, P. C.; Brouns, S. J. J.

CRISPR-Cas: Adapting to Change. Science (80-. )., 2017, 356 (6333).

https://doi.org/10.1126/science.aal5056.

[246] van der Oost, J.; Westra, E. R.; Jackson, R. N.; Wiedenheft, B. Unravelling the Struc tural and

Mechanistic Basis of CRISPR–Cas Systems. Nat. Rev. Microbiol., 2014, 12 (7), 479–492.

https://doi.org/10.1038/nrmicro3279.

[247] Barrangou, R.; Fremaux, C.; Deveau, H.; Richards, M.; Boyaval, P.; Moineau, S.; Romero, D. a;

Horvath, P. CRISPR Provides Acquired Resistance against Viruses in Prokaryotes. Science (80-. ).,


[248] Fineran, P. C. Resistance Is Not Futile: Bacterial ‘Innate’ and CRISPR -Cas ‘Adaptive’ Immune

Systems. Microbiology, 2019, 165 (8), 834–841. https://doi.org/10.1099/mic.0.000802.

[249] Stanley, S. Y.; Maxwell, K. L. Phage-Encoded Anti-CRISPR Defenses. Annu. Rev. Genet., 2018, 52

(1), 445–464. https://doi.org/10.1146/annurev-genet-120417-031321.

[250] Li, Y.; Bondy-Denomy, J. Anti-CRISPRs Go Viral: The Infection Biology of CRISPR-Cas

Inhibitors. Cell Host Microbe, 2021, 29 (5), 704–714. https://doi.org/10.1016/j.chom.2020.12.007.

[251] Bondy-Denomy, J.; Pawluk, A.; Maxwell, K. L.; Davidson, A. R. Bacteriophage Genes That

Inactivate the CRISPR/Cas Bacterial Immune System. Nature, 2013, 493 (7432), 429–432.

https://doi.org/10.1038/nature11723.

[252] Marino, N. D.; Zhang, J. Y.; Borges, A. L.; Sousa, A. A.; Leon, L. M.; Rauch, B. J.; Walton, R. T.;

Berry, J. D.; Joung, J. K.; Kleinstiver, B. P.; et al. Discovery of Widespread Type I and Type V

CRISPR-Cas Inhibitors. Science (80-. )., 2018, 362 (6411), 240–242.

https://doi.org/10.1126/science.aau5174.

[253] Hynes, A. P.; Rousseau, G. M.; Agudelo, D.; Goulet, A.; Amigues, B.; Loehr, J.; Romero, D. A.;

Fremaux, C.; Horvath, P.; Doyon, Y.; et al. Widespread Anti-CRISPR Proteins in Virulent

Bacteriophages Inhibit a Range of Cas9 Proteins. Nat. Commun., 2018, 9 (1), 1–10.

https://doi.org/10.1038/s41467-018-05092-w.

[254] Borges, A. L.; Zhang, J. Y.; Rollins, M. F.; Osuna, B. A.; Wiedenheft, B.; Bondy-denomy, J.

Bacteriophage Cooperation Suppresses CRISPR- Article Bacteriophage Cooperation Suppresses

CRISPR-Cas3 and Cas9 Immunity. Cell, 2018, 174 (4), 917-925.e10.

https://doi.org/10.1016/j.cell.2018.06.013.

[255] Chevallereau, A.; Meaden, S.; Fradet, O.; Gandon, S.; Houte, S. Van; Westra, E. R.; Chevallereau,

A.; Meaden, S.; Fradet, O.; Landsberger, M.; et al. Exploitation of the Cooperative Behaviors of

Anti- Article Exploitation of the Cooperative Behaviors of Anti-CRISPR Phages. Cell Host Microbe,

2020, 27 (2), 189-198.e6. https://doi.org/10.1016/j.chom.2019.12.004.

[256] Landsberger, M.; Gandon, S.; Meaden, S.; Rollie, C.; Chevallereau, A.; Chabas, H.; Buckling, A.;

Westra, E. R.; van Houte, S. Anti-CRISPR Phages Cooperate to Overcome CRISPR-Cas Immunity.


54

Cell, 2018, 174 (4), 908-916.e12. https://doi.org/10.1016/j.cell.2018.05.058.

[257] Bradde, S.; Mora, T.; Walczak, A. M. Cost and Benefits of Clustered Regularly Interspaced Short

Palindromic Repeats Spacer Acquisition. Philos. Trans. R. Soc. B Biol. Sci., 2019, 374 (1772).

https://doi.org/10.1098/rstb.2018.0095.

[258] Vale, P. F.; Lafforgue, G.; Gatchitch, F.; Gardan, R.; Moineau, S.; Gandon, S. Costs of CRISPR -

Cas-Mediated Resistance in Streptococcus Thermophilus. Proc. R. Soc. B Biol. Sci., 2015, 282

(1812), 20151270. https://doi.org/10.1098/rspb.2015.1270.

[259] Hesse, S.; Rajaure, M.; Wall, E.; Johnson, J.; Bliskovsky, V.; Gottesman, S.; Adhya, S. Phage

Resistance in Multidrug-Resistant Klebsiella Pneumoniae ST258 Evolves via Diverse Mutations

That Culminate in Impaired Adsorption. MBio, 2020, 11 (1), e02530-19.

https://doi.org/10.1128/mBio.02530-19.

[260] Welch, R. A. The Genus Escherichia. In The Prokaryotes; Dworkin, M., Falkow, S., Rosenberg, E.,

Schleifer, K.-H., Stackebrandt, E., Eds.; Springer, New York, NY, 2006; Vol. 6: Proteob, pp 60–71.

https://doi.org/110.1007/0-387-30746-x_9.

[261] Reshes, G.; Vanounou, S.; Fishov, I.; Feingold, M. Cell Shape Dynamics in Escherichia Coli.

Biophys. J., 2008, 94 (1), 251–264. https://doi.org/10.1529/biophysj.107.104398.

[262] Whitman, W. B.; Coleman, D. C.; Wiebe, W. J. Prokaryotes: The Unseen Majority . Proc. Natl. Acad.

Sci., 1998, 95 (12), 6578–6583. https://doi.org/10.1073/pnas.95.12.6578.

[263] Gordon, D. M.; Bauer, S.; Johnson, J. R. The Genetic Structure of Escherichia Coli Populations in

Primary and Secondary Habitats. Microbiology, 2002, 148 (5), 1513–1522.

https://doi.org/10.1099/00221287-148-5-1513.

[264] Savageau, M. A. Escherichia Coli Habitats, Cell Types, and Molecular Mechanisms of Gene Control.

Am. Nat., 1983, 122 (6), 732–744. https://doi.org/10.1086/284168.

[265] Kaper, J. B.; Nataro, J. P.; Mobley, H. L. T. Pathogenic Escherichia Coli. Nat. Rev. Microbiol., 2004,

2 (2), 123–140. https://doi.org/10.1038/nrmicro818.

[266] Tenaillon, O.; Skurnik, D.; Picard, B.; Denamur, E. The Population Genetics of Commensal

Escherichia Coli. Nat. Rev. Microbiol., 2010, 8 (3), 207–217. https://doi.org/10.1038/nrmicro2298.

[267] Denamur, E.; Picard, B.; Tenaillon, O. Population Genetics of Pathogenic Escherichia Coli. In

Bacterial Population Genetics in Infectious Disease; Robinson, D. A., Falush, D., Feil, E. J., Eds.;

John Wiley & Sons, Inc.: Hoboken, NJ, USA, 2010; pp 269–286.

https://doi.org/10.1002/9780470600122.ch14.

[268] Nolan, L. K.; Barnes, H. J.; Vaillancourt, J.; Abdul-aziz, T.; Logue, C. M. Colibacillosis. In Diseases

of Poultry; Swayne, D. E., Ed.; John Wiley & Sons, Inc., 2013; pp 751–805.

https://doi.org/10.1002/9781119421481.

[269] Kathayat, D.; Lokesh, D.; Ranjit, S.; Rajashekara, G. Avian Pathogenic Escherichia Coli (APEC):

An Overview of Virulence and Pathogenesis Factors, Zoonotic Potential, and Control Strategies.

Pathogens, 2021, 10 (4), 467. https://doi.org/10.3390/pathogens10040467.

[270] Lutful Kabir, S. M. Avian Colibacillosis and Salmonellosis: A Closer Look at Epidemiology,

Pathogenesis, Diagnosis, Control and Public Health Concerns. Int. J. Environ. Res. Public Health,

2010, 7 (1), 89–114. https://doi.org/10.3390/ijerph7010089.

[271] Dziva, F.; Stevens, M. P. Colibacillosis in Poultry: Unravelling the Molecular Basis of Virulence of

Avian Pathogenic Escherichia Coli in Their Natural Hosts. Avian Pathol., 2008, 37 (4), 355–366.

https://doi.org/10.1080/03079450802216652.

[272] Vandemaele, F.; Vereecken, M.; Derijcke, J.; Goddeeris, B. M. Incidence and Antiobiotic Resistance

of Pathogenic Escherichia Coli among Poultry in Belgium. Vet. Rec., 2002, 151 (12), 355–356.

https://doi.org/10.1136/vr.151.12.355.


55

[273] Ewers, C.; Antão, E. M.; Diehl, I.; Philipp, H. C.; Wieler, L. H. Intestine and Environment of the

Chicken as Reservoirs for Extraintestinal Pathogenic Escherichia Coli Strains with Zoonotic

Potential. Appl. Environ. Microbiol., 2009, 75 (1), 184–192. https://doi.org/10.1128/AEM.01324-08.

[274] Macklin, K. S.; Hess, J. B.; Bilgili, S. F.; Norton, R. A. Bacterial Levels of Pine Shavings and Sand

Used as Poultry Litter. J. Appl. Poult. Res., 2005, 14 (2), 238–245.

https://doi.org/10.1093/japr/14.2.238.

[275] Sarowska, J.; Futoma-Koloch, B.; Jama-Kmiecik, A.; Frej-Madrzak, M.; Ksiazczyk, M.; Bugla -

Ploskonska, G.; Choroszy-Krol, I. Virulence Factors, Prevalence and Potential Transmission of

Extraintestinal Pathogenic Escherichia Coli Isolated from Different Sources: Recent Reports. Gut

Pathog., 2019, 11, 10. https://doi.org/10.1186/s13099-019-0290-0.

[276] Rodriguez-Siek, K. E.; Giddings, C. W.; Doetkott, C.; Johnson, T. J.; Nolan, L. K. Characterizing

the APEC Pathotype. Vet. Res., 2005, 36 (2), 241–256. https://doi.org/10.1051/vetres:2004057.

[277] Vandemaele, F. J.; Mugasa, J. P.; Vandekerchove, D.; Goddeeris, B. M. Predominance of the PapGII

Allele with High Sequence Homology to That of Human Isolates among Avian Pathogenic

Escherichia Coli (APEC). Vet. Microbiol., 2003, 97 (3–4), 245–257.


[278] McCarter, J. D.; Stephens, D.; Shoemaker, K.; Rosenberg, S.; Kirsch, J. F.; Georgiou, G. Substrate

Specificity of the Escherichia Coli Outer Membrane Protease OmpT. J. Bacteriol., 2004, 186 (17),

5919–5925. https://doi.org/10.1128/JB.186.17.5919-5925.2004.

[279] Hejair, H. M. A.; Ma, J.; Zhu, Y.; Sun, M.; Dong, W.; Zhang, Y.; Pan, Z.; Zhang, W.; Yao, H. Role

of Outer Membrane Protein T in Pathogenicity of Avian Pathogenic Escherichia Coli. Res. Vet. Sci.,

2017, 115, 109–116. https://doi.org/10.1016/j.rvsc.2017.01.026.

[280] Torres, A. G.; Redford, P.; Welch, R. A.; Payne, S. M. TonB-Dependent Systems of Uropathogenic

Escherichia Coli: Aerobactin and Heme Tra nsport and TonB Are Required for Virulence in the

Mouse. Infect. Immun., 2001, 69 (10), 6179–6185. https://doi.org/10.1128/IAI.69.10.6179-

6185.2001.

[281] Sabri, M.; Caza, M.; Proulx, J.; Lymberopoulos, M. H.; Bree, A.; Moulin-Schouleur, M.; Curtiss, R.;

Dozois, C. M. Contribution of the SitABCD, MntH, and FeoB Metal Transporters to the Virulence

of Avian Pathogenic Escherichia Coli O78 Strain Χ7122. Infect. Immun., 2008, 76 (2), 601–611.

https://doi.org/10.1128/IAI.00789-07.

[282] Murase, K.; Martin, P.; Porcheron, G.; Houle, S.; Helloin, E.; Pénary, M.; Nougayrède, J. P.; Dozois,

C. M.; Hayashi, T.; Oswald, E. HlyF Produced by Extraintestinal Pathogenic Escherichia Coli Is a

Virulence Factor That Regulates Outer Membrane Vesicle Biogenesis. J. Infect. Dis., 2015, 212 (11),

856–865. https://doi.org/10.1093/infdis/jiv506.

[283] Janßen, T.; Schwarz, C.; Preikschat, P.; Voss, M.; Philipp, H. C.; Wieler, L. H. Virulence-Associated

Genes in Avian Pathogenic Escherichia Coli (APEC) Isolated from Internal Organs of Poultry

Having Died from Colibacillosis. Int. J. Med. Microbiol., 2001, 291 (5), 371–378.

https://doi.org/10.1078/1438-4221-00143.

[284] Li, Y.; Dai, J.; Zhuge, X.; Wang, H.; Hu, L.; Ren, J.; Chen, L.; Li, D.; Tang, F. Iron-Regulated Gene

IreA in Avian Pathogenic Escherichia Coli Participates in Adhesion and Stress-Resistance. BMC Vet.

Res., 2016, 12 (1), 167. https://doi.org/10.1186/s12917-016-0800-y.

[285] Johnson, T. J.; Wannemuehler, Y.; Doetkott, C.; Johnson, S. J.; Rosenberger, S. C.; Nolan, L. K.

Identification of Minimal Predictors of Avian Pathogenic Escherichia Coli Virulence for Use as a

Rapid Diagnostic Tool. J. Clin. Microbiol., 2008, 46 (12), 3987–3996.

https://doi.org/10.1128/JCM.00816-08.

[286] Schubert, S.; Rakin, A.; Karch, H.; Carniel, E. Prevalence of the “ High-Pathogenicity Island ” of

Yersinia Species among Escherichia Coli Strains That Are Pathogenic to Humans. Infect. Immun.,

1998, 66 (2), 480–485.


56

[287] Daines, D. A.; Wright, L. F.; Chaffin, D. O.; Rubens, C. E.; Silver, R. P. NeuD Plays a Role in the

Synthesis of Sialic Acid in Escherichia Coli K1. FEMS Microbiol. Lett., 2000, 189 (2), 281–284.

https://doi.org/10.1111/j.1574-6968.2000.tb09244.x.

[288] Zuo, J.; Tu, C.; Wang, Y.; Qi, K.; Hu, J.; Wang, Z.; Mi, R.; Yan huang; Chen, Z.; Han, X. The Role

of the Wzy Gene in Lipopolysaccharide Biosynthesis and Pathogenesis of Avian Pathogenic

Escherichia Coli. Microb. Pathog., 2019, 127, 296–303.

https://doi.org/10.1016/j.micpath.2018.12.021.

[289] Nolan, L. K.; Vaillancourt, J.; Barbieri, N. L.; Logue, C. M. Colibacillosis. In Diseases of Poultry;

Swayne, D. E., Boulianne, M., Logue, C. M., McDougald, L. R., Nair, V., Suarez, D. L., Wit, S.,

Grimes, T., Johnson, D., Kromm, M., et al., Eds.; Wiley, 2020; pp 770–830.

https://doi.org/10.1002/9781119371199.ch18.

[290] Johnson, T. J.; Wannemeuhler, Y. M.; Scaccianoce, J. A.; Johnson, S. J.; Nolan, L. K. Complete

DNA Sequence, Comparative Genomics, and Prevalence of an IncHI2 Plasmid Occurring among

Extraintestinal Pathogenic Escherichia Coli Isolates. Antimicrob. Agents Chemother., 2006, 50 (11),

3929–3933. https://doi.org/10.1128/AAC.00569-06.

[291] Johnson, T. J.; Johnson, S. J.; Nolan, L. K. Complete DNA Sequence of a ColBM Plasmid from

Avian Pathogenic Escherichia Coli Suggests That It Evolved from Closely Related ColV Virulence

Plasmids. J. Bacteriol., 2006, 188 (16), 5975–5983. https://doi.org/10.1128/JB.00204-06.

[292] Johnson, T. J.; Wannemuehler, Y. M.; Johnson, S. J.; Logue, C. M.; White, D. G.; Doetkott, C.;

Nolan, L. K. Plasmid Replicon Typing of Commensal and Pathogenic Escherichia Coli Isolates.

Appl. Environ. Microbiol., 2007, 73 (6), 1976–1983. https://doi.org/10.1128/AEM.02171-06.

[293] Johnson, T. J.; Siek, K. E.; Johnson, S. J.; Nolan, L. K. DNA Sequence of a ColV Plasmid and

Prevalence of Selected Plasmid-Encoded Virulence Genes among Avian Escherichia Coli Strains. J.


[294] Ginns, C. A.; Benham, M. L.; Adams, L. M.; Whithear, K. G.; Bettelheim, K. A.; Crabb, B. S.;

Browning, G. F. Colonization of the Respiratory Tract by a Virulent Strain of Avian Escherichia Coli

Requires Carriage of a Conjugative Plasmid. Infect. Immun., 2000, 68 (3), 1535–1541.

https://doi.org/10.1128/IAI.68.3.1535-1541.2000.

[295] Tivendale, K. A.; Allen, J. L.; Ginns, C. A.; Crabb, B. S.; Browning, G. F. Association of Iss and

IucA, but Not Tsh, with Plasmid-Mediated Virulence of Avian Pathogenic Escherichia Coli. Infect.

Immun., 2004, 72 (11), 6554–6560. https://doi.org/10.1128/IAI.72.11.6554.

[296] Mellata, M.; Dho-Moulin, M.; Dozois, C. M.; Curtiss, R.; Brown, P. K.; Arné, P.; Brée, A.;

Desautels, C.; Fairbrother, J. M. Role of Virulence Factors in Resistance of Avian Pathogenic

Escherichia Coli to Serum and in Pathogenicity. Infect. Immun., 2003, 71 (1), 536–540.

https://doi.org/10.1128/IAI.71.1.536-540.2003.

[297] Gibbs, P. S.; Maurer, A. J. J.; Nolan, B. L. K.; Wooleyc, R. E. Prediction of Chicken Embryo

Lethality with the Avian Escherichia Coli Traits Complement Resistance, Co licin V Production, and

Presence of the Increased Serum Survival Gene Cluster (Iss). Avian Dis., 2003, 47 (2), 370–379.

[298] Johnson, T. J.; Giddings, C. W.; Horne, S. M.; Gibbs, P. S.; Wooley, R. E.; Skyberg, J.; Olah, P.;

Kercher, R.; Sherwood, J. S.; Foley, S. L.; et al. Location of Increased Serum Survival Gene and

Selected Virulence Traits on a Conjugative R Plasmid in an Avian Escherichia Coli Isolate. Avian

Dis., 2002, 46 (2), 342–352. https://doi.org/10.1637/0005-2086(2002)046[0342:loissg]2.0.co;2.

[299] Mellata, M.; Ameiss, K.; Mo, H.; Curtiss, R. Characterization of the Contribution to Virulence of

Three Large Plasmids of Avian Pathogenic Escherichia Coli Χ7122 (O78:K80:H9). Infect. Immun.,

2010, 78 (4), 1528–1541. https://doi.org/10.1128/IAI.00981-09.

[300] Dozois, C. M.; Dho-Moulin, M.; Brée, A.; Fairbrother, J. M.; Desautels, C.; Curtiss, R. Relationship

between the Tsh Autotransporter and Pathogenicity of Avian Escherichia Coli and Localization and

Analysis of the Tsh Genetic Region. Infect. Immun., 2000, 68 (7), 4145–4154.

https://doi.org/10.1128/IAI.68.7.4145-4154.2000.


57

[301] Skyberg, J. A.; Johnson, T. J.; Johnson, J. R.; Clabots, C.; Logue, C. M.; Nolan, L. K. Acquisition of

Avian Pathogenic Escherichia Coli Plasmids by a Commensal E. Coli I solate Enhances Its Abilities

to Kill Chicken Embryos, Grow in Human Urine, and Colonize the Murine Kidney. Infect. Immun.,

2006, 74 (11), 6287–6292. https://doi.org/10.1128/IAI.00363-06.

[302] McPeake, S. J. W.; Smyth, J. A.; Ball, H. J. Characterisation of Avian Pathogenic Escherichia Coli

(APEC) Associated with Colisepticaemia Compared to Faecal Isolates from Healthy Birds. Vet.

Microbiol., 2005, 110 (3–4), 245–253. https://doi.org/10.1016/j.vetmic.2005.08.001.

[303] Kawano, M.; Yaguchi, K.; Osawa, R. Genotypic Analyses of Escherichia Coli Isolated from

Chickens with Colibacillosis and Apparently Healthy Chickens in Japan. Microbiol. Immunol., 2006,

50 (12), 961–966. https://doi.org/10.1111/j.1348-0421.2006.tb03872.x.

[304] Kauffmann, F. The Serology of the Coli Group. J Immunol., 1947, 57 (1), 71–100.

[305] DebRoy, C.; Fratamico, P. M.; Yan, X.; Baranzoni, G. M.; Liu, Y.; Needleman, D. S.; Tebbs, R.;

O’Connell, C. D.; Allred, A.; Swimley, M.; et al. Comparison of O-Antigen Gene Clusters of All O-

Serogroups of Escherichia Coli and Proposal for Adopting a New Nomenclature for O-Typing. PLoS

One, 2016, 11 (1), 1–13. https://doi.org/10.1371/journal.pone.0147434.

[306] Stenutz, R.; Weintraub, A.; Widmalm, G. The Structures of Escherichia Coli O-Polysaccharide

Antigens. FEMS Microbiol. Rev., 2006, 30 (3), 382–403. https://doi.org/10.1111/j.1574-

6976.2006.00016.x.

[307] Huja, S.; Oren, Y.; Trost, E.; Brzuszkiewicz, E.; Biran, D.; Blom, J.; Goesmann, A.; Gottschalk, G.;

Hacker, J.; Ron, E. Z.; et al. Genomic Avenue to Avian Colisepticemia. MBio, 2015, 6 (1), 1–13.


[308] Zhao, S.; Maurer, J. J.; Hubert, S.; De Villena, J. F.; McDermott, P. F.; Meng, J.; Ayers, S.; English,

L.; White, D. G. Antimicrobial Susceptibility and Molecular Characterization of Avian Pathogenic

Escherichia Coli Isolates. Vet. Microbiol., 2005, 107 (3–4), 215–224.


[309] Blanco, J.; Blanco, M.; Mora, A.; WH, J.; García, V.; Vá zquez, M. Serotypes of E Coli Isolated from

Septicaemic Chickens in Galicia (Northwest Spain). Vet. Microbiol., 1998, 61 (3), 229–235.

[310] Maiden, M. C. J.; Bygraves, J. A.; Feil, E.; Morelli, G.; Russell, J. E.; Urwin, R.; Zhang, Q.; Zhou,

J.; Zurth, K.; Caugant, D. A.; et al. Multilocus Sequence Typing: A Portable Approach to the

Identification of Clones within Populations of Pathogenic Microorganisms. Proc. Natl. Acad. Sci.,

1998, 95 (6), 3140–3145. https://doi.org/10.1073/pnas.95.6.3140.

[311] Wirth, T.; Falush, D.; Lan, R.; Colles, F.; Mensa, P.; Wieler, L. H.; Karch, H.; Reeves, P. R.; Maiden,

M. C. J.; Ochman, H.; et al. Sex and Virulence in Escherichia Coli: An Evolutionary Perspective.

Mol. Microbiol., 2006, 60 (5), 1136–1151. https://doi.org/10.1111/j.1365-2958.2006.05172.x.

[312] Mardis, E. R. The Impact of Next-Generation Sequencing Technology on Genetics. Trends Genet.,

2008, 24 (3), 133–141. https://doi.org/10.1016/j.tig.2007.12.007.

[313] Fratamico, P. M.; DebRoy, C.; Liu, Y.; Needleman, D. S.; Baranzoni, G. M.; Feng, P. Advances in

Molecular Serotyping and Subtyping of Escherichia Coli†. Front. Microbiol., 2016, 7 (644).


[314] Joensen, K. G.; Tetzschner, A. M. M.; Iguchi, A.; Aarestrup, F. M.; Scheutz, F. Rapid and Easy in

Silico Serotyping of Escherichia Coli Isolates by Use of Whole-Genome Sequencing Data. J. Clin.

Microbiol., 2015, 53 (8), 2410–2426. https://doi.org/10.1128/JCM.00008-15.

[315] Collett, S. R.; Smith, J. A.; Boulianne, M.; Owen, R. L.; Gingerich, E.; Singer, R. S.; Johnson, T. J.;

Hofacre, C. L.; Berghaus, R. D.; Stewart‐Brown, B. Principles of Disease Prevention, Diagnosis, and

Control. In Diseases of Poultry; Swayne, D. E., Boulianne, M., Logue, C. M., McDougald, L. R.,

Nair, V., Suarez, D. L., Wit, S., Grimes, T., Johnson, D., Kromm, M., et al., Eds.; Wiley, 2020; pp

1–78. https://doi.org/10.1002/9781119371199.ch1.


58

[316] Christensen, H.; Bachmeier, J.; Bisgaard, M. New Strategies to Prevent and Control Avian

Pathogenic Escherichia Coli (APEC). Avian Pathol., 2021, 50 (5), 370–381.

https://doi.org/10.1080/03079457.2020.1845300.

[317] Swelum, A. A.; Elbestawy, A. R.; El-Saadony, M. T.; Hussein, E. O. S.; Alhotan, R.; Suliman, G.

M.; Taha, A. E.; Ba-Awadh, H.; El-Tarabily, K. A.; Abd El-Hack, M. E. Ways to Minimize Bacterial

Infections, with Special Reference to Escherichia Coli, to Cope with the First -Week Mortality in

Chicks: An Updated Overview. Poult. Sci., 2021, 100 (5), 101039.

https://doi.org/10.1016/j.psj.2021.101039.

[318] Caekebeke, N.; Jonquiere, F. J.; Ringenier, M.; Tobias, T. J.; Postma, M.; van den Hoogen, A.;

Houben, M. A. M.; Velkers, F. C.; Sleeckx, N.; Stegeman, J. A.; et al. Comparing Farm Biosecurity

and Antimicrobial Use in High-Antimicrobial-Consuming Broiler and Pig Farms in the Belgian–

Dutch Border Region. Front. Vet. Sci., 2020, 7, 558455. https://doi.org/10.3389/fvets.2020.558455.

[319] U.S. Food and Drug Administration. Summary Report On Antimicrobials Sold or Distributed for Use

in Food-Producing Animals; 2020.

[320] Tang, K. L.; Caffrey, N. P.; Nóbrega, D. B.; Cork, S. C.; Ronksley, P. E.; Barkema, H. W.; Polachek,

A. J.; Ganshorn, H.; Sharma, N.; Kellner, J. D.; et al. Restricting the Use of Antibiotics in Food-

Producing Animals and Its Associations with Antibiotic Resistance in Food-Producing Animals and

Human Beings: A Systematic Review and Meta -Analysis. Lancet Planet. Heal., 2017, 1 (8), e316–

e327. https://doi.org/10.1016/S2542-5196(17)30141-9.

[321] Oliveira, A.; Sereno, R.; Azeredo, J. In Vivo Efficiency Evaluation of a Phage Cocktail in Controlling

Severe Colibacillosis in Confined Conditions and Experimental Poultry Houses. Vet. Microbiol.,

2010, 146 (3–4), 303–308. https://doi.org/10.1016/j.vetmic.2010.05.015.

[322] Li, H.; Ma, M. L.; Xie, H. J.; Kong, J. Biosafety Evaluation of Bacteriophages for Treatment of

Diarrhea Due to Intestinal Pathogen Escherichia Coli 3-2 Infection of Chickens. World J. Microbiol.

Biotechnol., 2012, 28 (1), 1–6. https://doi.org/10.1007/s11274-011-0784-5.

[323] Huff, W. E.; Huff, G. R.; Rath, N. C.; Donoghue, A. M. Method of Administration Affects the Ability

of Bacteriophage to Prevent Colibacillosis in 1-Day-Old Broiler Chickens. Poult. Sci., 2013, 92 (4),

930–934. https://doi.org/10.3382/ps.2012-02916.

[324] Naghizadeh, M.; Torshizi, M. A. K.; Rahimi, S.; Dalgaard, T. S. Synergistic Effect of Phage Therapy

Using a Cocktail Rather than a Single Phage in the Control of Severe Colibacillosis in Quails. Poult.

Sci., 2019, 98 (2), 653–663. https://doi.org/10.3382/ps/pey414.

[325] Huff, W. E.; Huff, G. R.; Rath, N. C.; Balog, J. M.; Xie, H.; Moore, P. A.; Donoghue, A. M.

Prevention of Escherichia Coli Respiratory Infection in Broiler Chickens with Bacteriophage

(SPR02). Poult. Sci., 2002, 81 (4), 437–441. https://doi.org/10.1093/ps/81.4.437.

[326] Wernicki, A.; Nowaczek, A.; Urban-Chmiel, R. Bacteriophage Therapy to Combat Bacterial

Infections in Poultry. Virol. J., 2017, 14 (1), 179. https://doi.org/10.1186/s12985-017-0849-7.

[327] El-Gohary, F. A.; Huff, W. E.; Huff, G. R.; Rath, N. C.; Zhou, Z. Y.; Donoghue, A. M. Environmental

Augmentation with Bacteriophage Prevents Colibacillosis in Broiler Chickens. Poult. Sci., 2014, 93

(11), 2788–2792. https://doi.org/10.3382/ps.2014-04282.

[328] Barrow, P.; Lovell, M.; Berchieri, A. Use of Lytic Bacteriophage for Control of Experimental

Escherichia Coli Septicemia and Meningitis in Chickens and Calves. Clin. Diagn. Lab. Immunol.,

1998, 5 (3), 294–298.

[329] Huff, W. E.; Huff, G. R.; Rath, N. C.; Balog, J. M.; Donoghue, A. M. Prevention of Escherichia Coli

Infection in Broiler Chickens with a Bacteriophage Aerosol Spray. Poult. Sci., 2002, 81 (10), 1486–

1491. https://doi.org/10.1093/ps/81.10.1486.

[330] Huff, W. E.; Huff, G. R.; Rath, N. C.; Balog, J. M.; Donoghue, A. M. Alternatives to Antibiotics:

Utilization of Bacteriophage to Treat Colibacillosis and Prevent Foodborne Pathogens. Poult. Sci.,

2005, 84 (4), 655–659. https://doi.org/10.1093/ps/84.4.655.


59

[331] Ryan, E. M.; Gorman, S. P.; Donnelly, R. F.; Gilmore, B. F. Recent Advances in Bacteriophage

Therapy: How Delivery Routes, Formulation, Concentration and Timing Influence the Success of

Phage Therapy. J. Pharm. Pharmacol., 2011, 63 (10), 1253–1264. https://doi.org/10.1111/j.2042-

7158.2011.01324.x.

[332] Pirnay, J.-P.; Ferry, T.; Resch, G. Recent Progress toward the Implementation of Phage Therapy in

Western Medicine. FEMS Microbiol. Rev., 2021, fuab040. https://doi.org/10.1093/femsre/fuab040.

[333] Huff, G.; Huff, W.; Rath, N.; Donoghue, A. Critical Evaluation of Bacteriophage to Prevent and

Treat Colibacillosis in Poultry. J. Ark. Acad. Sci., 2009, 63 (1), 93–98.

[334] Żbikowska, K.; Michalczuk, M.; Dolka, B. The Use of Bacteriophages in the Poultry Industry.

Animals, 2020, 10 (5), 872. https://doi.org/10.3390/ani10050872.

[335] Fernández, L.; Gutiérrez, D.; García, P.; Rodríguez, A. The Perfect Bacteriophage for Therapeutic

Applications—A Quick Guide. Antibiotics, 2019, 8 (3), 126.


Chapter 2: Scientific Aims

61


Chapter 2

Scientific Aims

Avian pathogenic E. coli (APEC) is one of the most important bacterial pathogens affecting

poultry. The emergence of multidrug-resistant pathogens has renewed the interest in alternative

treatment options, such as the therapeutic use of phages (phage therapy). Phages are viruses

that specifically infect bacteria and are the most abundant organism on Earth. In recent years,

an increasing amount of sequencing data becoming available has expanded our understanding

of phage diversity, but also revealed that we have only scratched the surface. A key first step

to develop a successful therapy is to build a collection of well-characterised candidate phages

targeting the pathogen of interest, as not all isolated phages meet the criteria for therapeutic

application. Nevertheless, there are still major problems with therapy, indicating our

understanding of the host-pathogen interaction is not well developed. We see that the dynamics

of interaction of the pathogen is variable and that a major concern of phage therapy is the

emergence of phage-resistant bacterial mutants. Therefore, the general aim of this PhD project

was to provide new insight into the diversity of coliphages, and to determine the coliphage-

bacterium interaction and population dynamics.


62

The specific aims of the project included:

1) Establish a well-characterised collection of lytic coliphages, using whole-genome

sequencing (WGS) analysis and bioinformatics tools (chapter 3.1).

2) Determine coliphage-host in vitro interactions and population dynamics in order to

have an in vitro model for better understanding of the pharmacodynamics and

pharmacokinetics (chapter 3.2).

3) Determine factors involved in APEC phage resistance, by generating in vitro resistant

combinations and determine the genetic background of resistance (chapter 3.3).

Chapter 3: Experimental Studies

63


Chapter 3

Experimental Studies

3.1 New insights into the biodiversity of coliphages in the intestine of poultry

3.2 Classification of in vitro phage-host population growth dynamics

3.3 Spontaneous phage resistance in avian pathogenic Escherichia coli

3.4 Schematic overview of the experimental studies and main findings


64


65

3.1

New insights into the biodiversity of coliphages in the intestine of

poultry

Patricia E. Sørensen1,2, Wim Van Den Broeck3, Kristoffer Kiil4, Dziuginta Jasinskyte5,

Arshnee Moodley5,6, An Garmyn1, Hanne Ingmer5, and Patrick Butaye1,2

1 Department of Pathology, Bacteriology and Poultry diseases, Ghent University, Belgium

2 Department of Biomedical Sciences, Ross University School of Veterinary Medicine, St. Kitts, West

Indies

3 Department of Morphology, Ghent University, Belgium

4 Department of Microbiology and Infection Control, Statens Serum Institut, Denmark

5 Department of Veterinary and Animal Sciences, University of Copenhagen, Denmark

6 CGIAR Antimicrobial Resistance Hub, International Livestock Research Institute, Nairobi, Kenya

Published in Scientific Reports 2020, 10 (15220)



66

Abstract

Despite phages’ ubiquitous presence and great importance in shaping microbial communities,

little is known about the diversity of specific phages in different ecological niches. Here, we

isolated, sequenced, and characterised 38 Escherichia coli-infecting phages (coliphages) from

poultry faeces to gain a better understanding of the coliphage diversity in the poultry intestine.

All phages belonged to either the Siphoviridae or Myoviridae family and their genomes ranged

between 44,324-173,384 bp, with a G+C content between 35.5-46.4%. Phylogenetic analysis

was performed based on single “marker” genes; the terminase large subunit, portal protein, and

exonucleases, as well as the full draft genomes. Single gene analysis resulted in six distinct

clusters. Only minor differences were observed between the different phylogenetic analyses,

including branch lengths and additional duplicate or triplicate subclustering. Cluster formation

was according to genome size, G+C content and phage subfamily. Phylogenetic analysis based

on the full genomes supported these clusters. Moreover, several of our Siphoviridae phages

might represent a novel unclassified phage genus. This study allowed for identification of

several novel coliphages and provides new insights to the coliphage diversity in the intestine

of poultry. Great diversity was observed amongst the phages, while they were isolated from an

otherwise similar ecosystem.


67

Introduction

Bacteriophages (phages) are viruses that have the ability to specifically infect bacteria. They

are estimated to be the most abundant form of life on Earth (~1031 organisms) and can be found

in almost every ecosystem, including soil, wastewater, sewage water, seawater and in and on

humans and animals [1–4]. Phages are thought to play essential roles in shaping the microbial

ecology, including driving the diversity of the bacterial communities [5]. As no single gene is

present in all phages, their taxonomic classification is based on host range, physical

characteristics, including size and morphology, genetic structure and composition, and overall

genome similarity [6, 7]. The phage classification scheme is regularly updated, refined and

approved by the International Committee on the Taxonomy of Viruses (ICTV) [8].

Furthermore, in recent years several genome-based phage taxonomy schemes have been

implemented [7, 9]. According to the National Center for Biotechnology information (NCBI),

as of February 2020, 9,238 complete phage genomes have been sequenced. However, despite

a continuously rising number of sequenced phage genomes, most of them remain unclassified

and poorly characterised. According to the ICTV, a phage genus can be defined as a group of

viruses with >50% nucleotide sequence similarity, which is distinct from viruses of other

genera. Moreover, defining characteristics can be determined for each genus, including average

genome length and number of coding sequences (CDSs), percentage of shared CDSs, and the

presence of specific signature genes in genus members [10].

Most phages that infect Escherichia coli, coliphages, belong the highly heterogeneous

Caudovirales order, which constitute ~96% of all known isolated phages [11]. This order

contains five families of tailed phages with dsDNA genomes: Myoviridae, Siphoviridae,

Podoviridae, Ackermannviridae and Herelleviridae [12]. According to ICTV taxonomy (data

of February 2020), these families comprise five, eleven, three, two, and five subfamilies,

respectively, and 87, 210, 48, three, and 15 genera, respectively. The currently analysed

Myoviridae coliphages belong to four subfamilies, including Ounavirinae, Peduovirinae,

Tevenvirinae, and Vequintavirinae, and 17 genera. Siphoviridae coliphages are found in only

two subfamilies: Guernseyvirinae and Tunavirinae, and in 13 genera. Podoviridae coliphages

belong to two subfamilies, the Autographivirinae and the Sepvirinae, and to 10 genera.

Ackermannviridae coliphages belong only to the Cvivirinae subfamily and the Kuttervirus

genus. To date, there have been no Herelleviridae coliphages isolated. To understand the

diversity, relationships, and dynamics among any group of phages, nucleotide sequence

information is needed [1]. For tailed phages, it has been reported that conserved genes such as


68

the terminase large subunit, the portal protein and major capsid protein (MCP), can be used as

phylogenetic markers for the diversity as well as their evolutionary relationship [1, 13].

Compared to their bacterial hosts, relatively few phages have been fully characterised

[14]. Besides, despite the phages’ significant role and ubiquitous presence in various areas,

little is known on the nature and extent of phage diversity in different ecosystems [3]. Recently,

there has been an interest in the diversity of coliphages [1, 15, 16]. Here, we performed a

detailed genome-based characterisation and phylogenetic analysis of 38 fully sequenced

coliphages, all isolated from a single, relatively unexplored environmental source: poultry

faecal material.

Methods

Phage isolation and propagation

The phages were isolated from poultry faecal material, collected randomly from 27 poultry

houses in Belgium in 2013. Phages were propagated according to Adams (1959) and Bonilla

et al. (2016) with minor modifications [17, 18]. Briefly, the samples (5g) were emulsified in

Lysogeny broth(LB) broth (Miller)(Sigma-Aldrich, Saint Louis, MO, USA). The decanted

supernatant obtained from each emulsion was enriched by the addition of two early -log phase

host bacteria, E. coli K-12 derived laboratory strains C600 [19] or K514 [20], a non-restricting,

modifying derivative of strain C600. Suspensions were incubated overnight at 37°C, with

shaking (120 rpm) and were then centrifuged at 4,000 rpm for 30 min to pellet the cellular

debris. The supernatant containing the phage(s) was centrifuged and filtered using a 0.45 µm

membrane filter followed by a 0.2 µm Minisart Filter (Fisher Scientific, Waltham, MA, USA).

The enriched phage suspensions were enumerated and tested for lytic activity on the host

bacteria using the double-layer agar (DLA) technique [17, 21, 22]. Briefly, phage suspensions

were serial diluted and spotted on an overlay of the respective host bacteria on solid LB medium

supplemented with 0.8% agar and 0.5 mM CaCl2. A clear zone in the plate, a plaque, resulting

from the lysis of host bacterial cells, indicated the presence of virulent coliphage(s). Samples

with lytic activity against the indicator strain were further processed for single phage plaque

isolation, including three rounds of plaque purification, and propagation. All phage lysates

were stored at 4°C until required.

Phage morphological analysis


69

The morphology of unique coliphages (≤ 95% nucleotide similarity) isolated in this study was

investigated using transmission electron microscopy (TEM). Phage suspension was applied to

the surface of Formvar carbon-coated grids, the phages were fixed using paraformaldehyde

(PFA) (4% w/v), washed, and negatively stained with UrAC (1% w/v). After drying, grids were

examined using a JEM-1400 Plus transmission electron microscope (JEOL, Benelux).

Genomic DNA extraction and sequencing

DNA extraction from phage lysates was performed using DNeasy Blood & Tissue Kit (Qiagen,

Hilden, Germany) as previously described [23]. The DNA concentration and quality was

assessed using the NanoDrop (Thermo Scientific, Roskilde, Denmark) and Qubit fluorometer

(Thermo Scientific, Roskilde, Denmark) according to the manufacturer’s instructions.

Preparation of paired-end 2 x 250 bp sequencing libraries was done using the Nextera XT Kit

(Illumina, San Diego, USA) with adaptations for phage genomes as shown elsewhere [24] and

sequenced on the Illumina MiSeq platform using MiSeq Reagent Kit v2 (500-cycles) and

manufacturer’s instructions, yielding a total of 16,270-237,128 paired end reads for each phage

lysate. Read-pair contigs were generated for each MiSeq cluster prior to assembly.

Phage genome sequence analysis and annotation

FastQC software (https://www.bioinformatics.babraham.ac.uk/projects/fastqc/), v0.11.3, was

used for quality control validation of the raw reads sequence data. Low-quality sequences were

excluded from further analysis. The raw reads were trimmed for quality, adaptor sequences

were removed using default parameters. The sequence reads were de novo assembled using

QIAGEN Bioinformatics CLC Genomic Workbench, v11.0.1, using default settings, with

minimum contig length changed to 250 bp. An overview of assembly statistics is provided in

Supplementary Table S3. The assembled draft phage sequences were compared with phage

homologues from the NCBI nucleotide database (https://blast.ncbi.nlm.nih.gov/Blast.cgi)

using the Basic Local Alignment Search Tool (BLAST) software [25], and from the custom

PHAge Search Tool Enhanced Release (PHASTER) phage database [26]. Newly assembled

phage sequences were compared using both BLAST and PHASTER to identify unique and

identical (>95% nucleotide sequence similarity) phages. Assembled contigs were submitted to

the ResFinder database, v3.2 [27] and the VirulenceFinder database, v2.0 [28] to identify any

acquired antimicrobial resistance and virulence associated genes, respectively. By default,

selected threshold for %ID was 90% and 60% for minimum length. All 15 databases for

https://www.bioinformatics.babraham.ac.uk/projects/fastqc/


70

antimicrobial resistance genes were selected. The taxonomic group of E. coli bacteria was

selected for VirulenceFinder. PHASTER and The Rapid Annotation using Subsystem

Technology (RAST) server and the SEED viewer, v2.0, were used for identification of CDSs

and initial annotation of the phage genomes, including identification of the phage terminase

large subunit [29]. Gene function of genes defined as “hypothetical protein” was predicted by

comparison to homologue genes with defined functions in other related phage genomes. The

G+C content of the phages was calculated using the SEED viewer.

Phage phylogeny and taxonomy

Multiple genome alignment of the WGS sequences was performed using Applied Maths

BioNumerics software, v7.6. According to the ICTV taxonomy guidelines, the 38 coliphages

were classified into phage family, subfamily and genera based on nucleotide similarity to

known Siphoviridae and Myoviridae coliphages. Known reference coliphages included were

limited to isolates with complete genomes found in the ICTV database and the NCBI database

(data of November 2019).

Phage diversity

Phylogenetic trees based on the phage whole genome sequences were constructed using R,

v3.5.1 [30], for comparison of the coliphages with published Siphoviridae or Myoviridae

reference phage genomes (accessed f rom the NCBI database). The trees were constructed using

unweighted pair group method with arithmetic mean (UPGMA) from a distance matrix of

binary distances calculated from either, gene presence/absence within the full genomes of the

phages determined using Roary v3.12.0 [31], Prokka, v1.13.7 [32] and prodigal, v2.6.3 [33],

or Kmer presence/absence (using 10 and 21mer) based on de novo assembled contigs, as

calculated using a python script (Supplementary Script kmer.py).

For phylogenetic analysis based on single marker gene, phage gene sequences were aligned

using Clustal X, v2.1 [34]. A maximum likelihood phylogenetic trees (unrooted) was

constructed and supported by bootstrap analysis (inferred from 1000 replicates) with default

substitution model (Tamura-Nei model) to assess the diversity of the coliphages using the

phylogenetic and molecular evolutionary genetics analyses (MEGA) software, vX [35]. The

phylogenetic trees were based on the nucleotide sequences of the CDSs of the following genes:


71

phage terminase large subunit, phage portal protein, or phage exonucleases. The reference

genomes included, represented the best matching published sequences to the phages in this

study (selected based on the BLAST max score) and core reference genomes for comparison.

The degree of topological and branch length agreement between the different phylogenetic

methods and between the three marker genes was investigated using the R packages Analysis

of Phylogenetics and Evolution (ape) [36] and phangorn [37].

Phage comparative genomics

A more detailed analysis of the most closely related coliphage genomes was carried out.

Genomes were re-annotated using Prokka and pan genome analysis was carried out with Roary

using script “roary -e -n -s -p 20 -i 90 *.gff”, including identification of core genome, including

core and softcore genes, and accessory genome, including shell and cloud genes.

To investigate the level of synteny and genomic rearrangement, whole genome alignment and

comparison of coliphages and related reference phages from each cluster or subcluster were

performed with the Mauve software using progressiveMauve [38] with default parameters. No

more than 19 of the most related phages were included in each comparison for simplification.

Relatedness of the phages were based on percentage of nucleotide similarity and number of

shared core genes. Reference genomes were included for annotation references.

Results

Phage isolation

In this study, 38 coliphages were isolated from poultry faecal samples collected from 27

Belgian poultry farms located in five different regions, including West Flanders, East Flanders,

Antwerp, and Limburg. Between one and seven phages were isolated from each farm using E.

coli C600 or K514 as host strain.

Phage morphological analysis

Based on a sequencing cut-off value of ≤95% nucleotide similarity, 18 coliphages were

selected and subjected to TEM to determine phage morphology and confirm phage


72

classification. Based on the morphological features, the phages were classified into the

Caudovirales order and either the Siphoviridae family or the Myoviridae family. Analysis of

the isolated Siphoviridae phages showed a long flexible non-contractile tail with a length

varying between ~100 nm and ~200 nm and icosahedral heads with widths ranging from ~52

nm and ~77 nm (Fig. 1a-h).

Figure 1 | Negative staining electron microscopy images of Siphoviridae and Myoviridae coliphages. Siphoviridae

phages: a) Phage 17. b) Phage 53. c) Phage 54. d) Phage 61. e) Phage 70. f) Phage 74. g) Phage 76. h) Phage 77.

Myoviridae phages: i) Phage 10. j) Phage 11. k) Phage 15. l) Phage 18. m) Phage 30. n) Phage 55. o) Phage 60.

p) Phage 62 q) Phage 78. r) Phage 79. The black bars represent 100 nm.


73

Among the isolated Myoviridae phages a long straight contractile tail was observed with a

tailed length varied between ~100 nm and ~120 nm, head widths ranging from ~65 nm to ~84

nm, and head lengths from ~60 nm to ~110 nm (Fig. 1i-r). Taxonomic classification of each of

the coliphages is shown in Table 1.

Phage genome sequence analysis and annotation

All 38 coliphages isolated in this study were characterised based on WGS data. An overview

of the genomic characteristics and properties are listed in Table 1. According to FastQC

parameters, good quality of the raw sequence data for all phages was confirmed. The phage

genomes ranged in size between 44,324 bp to 173,384 bp, with a G+C content between 35.5%

and 46.4%. Genomes smaller than 90,000 bp had a G+C content between 38.9% and 46.4%,

whereas the larger genomes had a G+C content of 35.5-38%. For each coliphage, 72 to 275

putative CDSs were identified using both automatic and manual annotation. CDSs encoding

the phage terminase small subunit, the phage terminase large subunit, the phage portal protein,

and phage capsid and scaffold proteins were identified within all 38 coliphage genome

sequences. They presented the same conserved genome structure with a general gene order: the

terminase small subunit upstream from the terminase large subunit, the phage portal protein

and two genes encoding phage capsid and scaffold proteins. In general, one phage terminase

small subunit, one phage portal protein, and up to four phage capsid and scaffold proteins were

found within each of the phage genomes. Besides, phage exonucleases were identified in all

phage genomes. For each phage, one to three CDSs for exonucleases were found. No gene

encoding for an integrase was found, indicating that these phages are strictly virulent/lytic

phages. No known acquired antimicrobial resistance or virulence genes were detected in any

of the 38 phage genomes.

Phage phylogeny and taxonomy

Taxonomic classification of the 38 coliphages was performed through multiple WGS genome

comparisons. These coliphages included 27 (71%) Siphoviridae coliphages and 11 (29%)

Myoviridae coliphages. The Siphoviridae phages were compared with 146 published phages

from this family. The Myoviridae phages were compared with 171 published Myoviridae

phages. According to ICTV guidelines, phage family, subfamily and genus were predicted

based on genome similarity. Results are shown in Table 1.


74

Ta

ble

1 | C

hara

cte

rist

ics

of

the 3

8 E

. co

li-i

nfe

cti

ng p

hages

invest

igate

d in

th

is s

tud

y

Ph

age

nam

e

Reg

-

ion

Farm

ID

E. co

li

ho

st

Gen

om

e

size

(b

p)

%

G+

C

#

CD

Ss

Rel

ate

d r

ef.

ph

ag

e

Ph

age

fam

ily

Ph

age

sub

fam

ily

P

hage

gen

us

Ph

age

clu

ster

**

WG

S T

LS

P

P E

xo

Phag

e 47

O

VL

1

5

C6

00

5

10

63

43

.6

83

G2

9-2

S

iph

o

Tunavi

rinae

Han

rive

rvir

us

A1

A1

A1

A1

Phag

e 48

W

VL

1

2

C6

00

5

10

31

43

.7

85

Hen

u8

Sip

ho

T

unavi

rinae

Han

rive

rvir

us

A1

A1

A1

A1

Phag

e 53

W

VL

1

0

K5

14

5

08

35

44

.2

87

G2

9-2

S

iph

o

Tunavi

rinae

Han

rive

rvir

us

A1

A1

A1

A1

Phag

e 54

W

VL

1

4

K5

14

5

26

02

43

.5

88

Hen

u8

Sip

ho

T

unavi

rinae

Han

rive

rvir

us

A1

A1

A1

A1

Phag

e 59

L

IM

22

K

51

4

51

70

2

43

.6

85

G2

9-2

S

iph

o

Tunavi

rinae

Han

rive

rvir

us

A1

A1

A1

A1

Phag

e 63

O

VL

1

9

K5

14

4

91

32

44

.0

79

G2

9-2

S

iph

o

Tunavi

rinae

Han

rive

rvir

us

A1

A2

A1

A1

Phag

e 64

O

VL

1

9

K5

14

5

13

52

43

.7

85

G2

9-2

S

iph

o

Tunavi

rinae

Han

rive

rvir

us

A1

A1

A1

A1

Phag

e 65

O

VL

1

9

K5

14

5

10

31

43

.6

83

G2

9-2

S

iph

o

Tunavi

rinae

Han

rive

rvir

us

A1

A1

A1

A1

Phag

e 68

O

VL

2

1

K5

14

5

12

91

43

.7

84

G2

9-2

S

iph

o

Tunavi

rinae

Han

rive

rvir

us

A1

A1

A1

A1

Phag

e 71

O

VL

19

K

514

51

44

6

43.6

85

G29-2

Sip

ho

T

unavi

rinae

Hanri

verv

irus

A1

A1

A1

A1

Phag

e 72

O

VL

17

K

514

51

28

4

43.7

84

G29-2

Sip

ho

T

unavi

rinae

Hanri

verv

irus

A1

A1

A1

A1

Phag

e 75

O

VL

20

K

514

50

44

5

44.1

88

G29-2

Sip

ho

T

unavi

rinae

Hanri

verv

irus

A1

A1

A1

A1

Phag

e 77

A

NT

6

K514

51

07

3

44.0

85

G29-2

Sip

ho

T

unavi

rinae

Hanri

verv

irus

A1

A1

A1

A1

Phag

e 8

V

BR

27

C

600

51

03

1

43.6

83

pS

f-1

Sip

ho

T

unavi

rinae

Hanri

verv

irus

A1

A1

A1

A1

Phag

e 28

O

VL

19

K

514

52

97

0

44.4

87

SE

Cphi2

7

Sip

ho

T

unavi

rinae

Sw

anvi

rus*

A

2

A2

A2

A2

Phag

e 56_1

A

NT

1

K514

52

71

6

44.5

86

Eco

S-9

5

Sip

ho

T

unavi

rinae

Sw

anvi

rus*

A

2

A2

A2

A2

Phag

e 76

W

VL

11

K

514

51

90

5

45.0

93

SE

Cphi2

7

Sip

ho

T

unavi

rinae

Sw

anvi

rus*

A

2

A4

A2

A2

Phag

e 80

O

VL

18

K

514

52

70

3

44.5

88

Eco

S-9

5

Sip

ho

T

unavi

rinae

Sw

anvi

rus*

A

2

A2

A2

A2

Phag

e 52

W

VL

13

K

514

53

01

8

45.9

90

Jahat

MG

145

Sip

ho

T

unavi

rinae

New

gen

us

A3

A3

A3

A3

Phag

e 56_2

A

NT

1

K514

50

82

9

45.7

87

Jahat

MG

145

Sip

ho

T

unavi

rinae

New

gen

us

A3

A3

A3

A3


75

Ta

ble

1 | C

on

tin

ued

Ph

age

nam

e R

eg-

ion

F

arm

ID

E

. co

li

host

G

eno

me

size

(b

p)

%

G+

C

#

CD

Ss

Rel

ate

d r

ef.

ph

age

Ph

ag

e fa

mil

y

Ph

age

sub

fam

ily

P

hage

gen

us

Ph

age

clu

ster

**

WG

S T

LS

P

P

Exo

Phag

e 69

W

VL

14

K

514

62

38

4

46.3

112

Jahat

MG

145

Sip

ho

T

unavi

rinae

New

gen

us

A3

A3

A3

A3

Phag

e 17

O

VL

19

K

514

45

94

8

44.5

73

CE

B_E

C3a

Sip

ho

T

unavi

rinae

Rtp

viru

s B

B

B

B

Phag

e 58

A

NT

4

K514

45

38

7

44.3

73

CE

B_E

C3a

Sip

ho

T

unavi

rinae

Rtp

viru

s B

B

B

B

Phag

e 70

L

IM

25

K

514

44

53

9

44.8

72

CE

B_E

C3a

Sip

ho

T

unavi

rinae

Rtp

viru

s B

B

B

B

Phag

e 73

O

VL

17

K

514

46

93

8

44.3

76

CE

B_E

C3a

Sip

ho

T

unavi

rinae

Rtp

viru

s B

B

B

B

Phag

e 74

O

VL

21

K

514

46

68

3

44.6

77

CE

B_E

C3a

Sip

ho

T

unavi

rinae

Rtp

viru

s B

B

B

B

Phag

e 61

A

NT

3

K514

10

98

66

39.2

164

T5

Sip

ho

N

/A

Teq

uin

tavi

rus

C

C

C

C

Phag

e 60

L

IM

26

K

514

86

23

7

39.3

127

Alf

5

Myo

O

unavi

rinae

Fel

ixoun

avi

rus

D

D

D

D

Phag

e 62

V

BR

7

K514

87

87

1

39.0

128

Alf

5

Myo

O

unavi

rinae

Fel

ixoun

avi

rus

D

D

D

D

Phag

e 66

A

NT

2

K514

90

19

6

39.0

137

Alf

5

Myo

O

unavi

rinae

Fel

ixoun

avi

rus

D

D

D

D

Phag

e 78

V

BR

8

K

51

4

89

90

0

39

.1

13

0

Alf

5

Myo

O

unavi

rinae

Fel

ixou

na

viru

s D

D

D

D

Phag

e 79

L

IM

23

K

514

89

66

3

39.0

135

AY

O145

A

Myo

O

unavi

rinae

Fel

ixoun

avi

rus

D

D

D

D

Phag

e 15

O

VL

2

1

K5

14

1

69

58

6

37

.7

26

9

MM

02

M

yo

Tev

envi

rinae

Mosi

gvi

rus

E

E

E

E

Phag

e 18

O

VL

1

9

K5

14

1

69

86

8

37

.7

27

1

MM

02

M

yo

Tev

envi

rinae

Mosi

gvi

rus

E

E

E

E

Phag

e 30

W

VL

9

K

51

4

17

33

84

38

.0

27

5

O1

57

tp

3

Myo

T

even

viri

nae

Mosi

gvi

rus

E

E

E

E

Phag

e 10

A

NT

5

K

51

4

16

89

52

35

.5

26

8

YU

EE

L0

1

Myo

T

even

viri

nae

Teq

uatr

ovi

rus

F

F

F

F

Phag

e 11

O

VL

1

6

K5

14

1

71

37

0

35

.5

26

9

fFiE

co0

6

Myo

T

even

viri

nae

Teq

uatr

ovi

rus

F

F

F

F

Phag

e 55

L

IM

24

K

51

4

16

99

53

5

35

.6

27

5

Ph

age

T4

Myo

T

even

viri

nae

Teq

uatr

ovi

rus

F

F

F

F

AN

T =

An

twerp

, VB

R =

Fle

mis

h (V

laa

ms)

Bra

bant,

WV

L =

West

Fla

nd

ers,

OV

L =

Ea

st F

lan

der

s, a

nd

LIM

= L

imb

urg

. S

ipho =

Sip

ho

viri

da

e, M

yo

= M

yovir

idae

N/A

= N

o s

ub

fam

ily

is d

efi

ned a

cco

rdin

g to

th

e I

nte

rnati

onal

Co

mm

itte

e o

n T

axonom

y V

iru

ses

(IC

TV

)

* S

wa

nvir

us

gen

us

is n

ot y

et acce

pte

d in

th

e I

CT

V d

ata

base

[3

9].

**

Ph

age c

lust

er b

ase

d o

n w

hole

gen

om

e se

quence

(WG

S),

or th

e s

ingle

sig

na

ture

gen

es:

term

inase

larg

e s

ub

unit

(T

LS

), p

ort

al p

rote

in (P

P),

or ex

on

ucle

ase

s (E

xo)


76

All Siphoviridae phages belonged to the Tunavirinae subfamily, except for Phage 61. This

phage was predicted to belong to the Tequintavirus genus, which do not have any ICTV

subfamily. The 10 phages, Phage 8, 53, 54, 63, 65, 68, 69, 71, 72, and 75 all belonged to the

Hanrivervirus genus. Three phages belonged to the Rtpvirus genus, including Phage 17, 70,

and 73. No existing ICTV genus could be assigned to the remaining 13 coliphages. Phage 28,

56_1 and 76 could be assigned to the same unknown genus. Phage 58 and Phage 74 were found

to be in the same genus. Phage 47, 48, 59, 64, and 77 were predicted to belong to the same

genus. Phage 52, 56_2, and 80 were predicted to belong to the same genus. The 11 Myoviridae

belonged either to the Tevenvirinae or the Ounavirinae subfamily. Tevenvirinae phages

included the six phages: Phage 10, 11, 15, 18, 30 and 55. Phage 10, 11 and 55 belonged to the

Tequatrovirus genus, and Phage 18 and 30 belonged to the Mosigvirus genus. Ounavirinae

phages included the remaining five phages; Phage 60, 62, 66, 78 and 79. All phages belonged

to the Felixounavirus genus.

Phage diversity

To investigate the diversity of the coliphages, phage genomes were first clustered based on

whole genome sequence. A total of 173 Siphoviridae and 182 Myoviridae coliphage genomes

were included. Characteristics of selected reference genomes are listed in Supplementary Table

S1. Siphoviridae phages isolated in this study were found in five different (sub)clusters, cluster

A1-3, B and C, with a cut-off value of 0.82 (Fig. 2). Cluster A was divided into three

subclusters. Fourteen of our phages, formed subcluster A1 together with the three pSf-1-like

reference phages from the NCBI database. Phage 80, 28, 56_1, and 76 formed subcluster A2

with the three Swan01-like reference phages. Phages 69, 52, and 56_2 formed subcluster A3

with phage Jahat_MG145. Phage 73, 70, 17, 58 and 74 formed cluster B without any known

reference phages. Phage 61 was placed in cluster C with 13 T5-like reference phages. For the

Myoviridae phages, the resulting phylogeny placed phages isolated in this study in three

different clusters with a cut-off height of 0.52 (Fig. 3). Phages 62, 78, 66, 60 and 79 formed

cluster D with Felix01-like reference phage Alf5. Phage 30, 15 and 18 formed cluster E with

19 T4-like reference phages (cut-off height of 0.36). Phages 55, 11 and 10 were placed in

cluster F with 57 reference phages. At the cut-off height of 0.39 Phage 55 was found in a

different subcluster than Phage 10 and 11.


77

Fig

ure 2

| P

hy

logenetic a

naly

sis

of

Sip

hovir

idae c

olip

hages

base

d o

n W

GS

seq

uence.

Ph

ages

iso

late

d in

th

is s

tudy a

re h

igh

ligh

ted. E

ach c

olo

ur

rep

rese

nts

a c

lust

er:

Clu

ster A

(b

lue),

clu

ster B

(gre

en

), a

nd

clu

ster C

(re

d).

Clu

ster A

su

bclu

ster

s in

clu

de A

1 (ligh

t b

lue),

A2

(b

lue),

an

d A

3 (d

ark

blu

e). D

ista

nce

matr

ices a

nd c

lust

ering a

re

ba

sed o

n k

mer le

ngth

= 1

0.


78

Fig

ure 3

| P

hy

logenetic a

nal

ysi

s o

f M

yo

viri

dae c

olip

hages

base

d o

n W

GS

seq

uence

. P

hages

iso

late

d in

th

is s

tudy a

re h

igh

ligh

ted. E

ach c

olo

ur

rep

rese

nts

a c

lust

er:

Clu

ster D

(o

ran

ge),

clu

ster E

(p

urp

le),

and

clu

ster F

(b

row

n).

Dis

tan

ce m

atr

ices

and c

lust

eri

ng a

re b

ase

d o

n k

mer le

ngth

= 1

0


79

Coliphages were further assessed based on the presence/absence of families of orthologues

genes in their pan genome. Similar clusters were observed with only minor changes. For the

Siphoviridae phage analysis, 5227 gene groups were included (Supplementary Fig. S1). The

resulting phylogenetic analysis placed phages isolated in this study in the same five clusters,

cluster A1-3, B and C, with a cut-off height of 0.81 (Supplementary Fig. S2). One additional

reference phage was found in cluster B and C, including the T1-like reference phage

CEB_EC3a and the T5-like reference phage EPS7, respectively. For the Myoviridae phage

analysis, 9420 gene groups were included (Supplementary Fig. S3). The resulting phylogeny

placed phages isolated in this study in the same three clusters, cluster D, E, and F, with a cut-

off height of 0.58 (Supplementary Fig. S4). For cluster D, additionally 13 Felix01-like

reference phages were found. In contrast to the WGS-based analysis, at a cut-off height of 0.39,

all cluster F phages isolated in this study, were found in one subcluster with 10 T4 -like

reference phages. The degree of topological and branch length agreement between the different

phylogenetic methods were compared (Supplementary Table S2).

The coliphage diversity was further assessed based on three phage marker genes: the terminase

large subunit and phage portal protein, and the phage exonuclease. Selected gene sequences

from known phages were included for reference. Results are summarised in Table 1. For all

three marker genes, cluster formation was in accordance with resulting clusters of the pan

genome- and WGS-based phylogeny, cluster A-F, only with minor differences. Results based

on the terminase large subunit analysis are shown below (Fig. 4).

For cluster A, all coliphages isolated in this study were found within same subclusters as for

the WGS-based phylogeny except for Phage 63, which was found in the A2 subcluster instead

of A1. Analysis based on the phage portal protein resulted in the division of our A2 subcluster

phages into two groups: Phage 56_1, 80 and 28 in one group and Phage 76 in the other group

(Supplementary Fig. S5). Analysis based on the exonuclease resulted in multiple clusters of

cluster C and F, as phages from these clusters encoded two or two-three exonuclease genes,

respectively (Supplementary Fig. S6). Comparison of the cluster construction of the three

single genes analysis showed only minor topological and branch length differences

(Supplementary Table S2). Moreover, cluster construction was in accordance with phage

subfamily defined based on the whole genome. Siphoviridae phages from cluster A and B

belonged to the Tunavirinae subfamily, and Siphoviridae phages form cluster C had no defined

ICTV subfamily. Myoviridae phages from cluster D belonged to the Ounavirinae subfamily,

and Myoviridae phages from cluster E and F belonged to the Tevenvirinae subfamily.


80

Figure 4 | Maximum likelihood tree based on the nucleotide sequences of the phage terminase large subunit. The

analysis resulted in six clusters: A-F, according to phage family and subfamily. Cluster A and B: Siphoviridae,

Tunavirinae, cluster C: Siphoviridae and Tequintavirus genus, cluster D: Myoviridae, Ounavirinae, and cluster E

and F: Myoviridae, Tevenvirinae. Cluster A was divided into three subclusters: A1, A2 and A3. The tree was

constructed using the MEGA X software [35]. The percent of data coverage for internal nodes is indicated. The

scale bar indicates the number of nucleotide sequence substitutions per site. The analysis included 62 nucleotide

sequences, including 24 reference phages listed in Supplementary Table S1 for comparison.

Phage comparative genomics

Pan genome analysis of Siphoviridae and Myoviridae phages isolated in this study revealed

that neither of the two groups had any core genes. Analysis of coliphage genomes from each

of the six clusters, A-F, identified core genes (core and softcore) and accessory genes (shell

and cloud). As cluster A phages had only five core genes (2% of the total genome), analysis of

subclusters, A1, A2 and A3, were performed additionally. Results are summarised in Table 2.


81

The pan genome included between 81 and 333 genes, and core genes constituted between 22%

and 73% of the pan genome. The level of synteny and genomic rearrangement within each

cluster or subcluster of related phages was assessed by genome comparison (Table 2). Eight

comparisons were performed, corresponding to the eight (sub)clusters, A1, A2, A3, B, C, D,

E, and F resulting from the phage diversity analysis above (Supplementary Fig. S7 -14).

Genome comparison of the phages resulted in identification of local collinear blocks (LCBs),

indicating homologues DNA regions shared by two or more genomes without sequence

rearrangements. The LCBs comprised different modules of genes with different functions,

including modules for DNA packaging, structural proteins, head and tail morphogenesis, and

host cell lysis. Several modules comprised only hypothetical proteins with unknown function.

The average level of conservation varied between the different type of genes.

Genes encoding the terminase large and small subunit, the MCP, DNA primase, portal protein,

recombinase, specific tail protein and holin were the most conserved genes between all phages,

whereas genes with the lowest level of conservation included, tail fiber proteins, tail tape

measure proteins and HNH homing endonucleases. Hypothetical proteins were found with

large variation in level of conservation. Each phage genome comprised between four and 17

LCBs. Genome comparison subcluster A1, A2 and A3 phages identified 16, seven and four

LCBs, respectively. All phages in each cluster comprised all LCBs. All cluster B phages

comprised all six LCBs. For the cluster C phages, between six and 10 LCBs were identified

for each phage. Phage 61 comprised all nine regions. Variation in number of LCBs was due to

a variable repeat region comprising multiple LCBs, which was found only in some of the

cluster C phages. For the cluster D comparison, 14-17 different LCBs were identified for each

phage. Variation in number of LCBs was due to four different small variable regions of which

some of all were missing in the majority of the phages. Phages isolated in this study, including

Phages 79, 78, 60, 66, and 62, comprised 17, 17, 16, 15, and 14 LCBs, respectively.

Comparison of phages belonging to cluster E identified 18 LCBs. All phages lacked one or

both of the same two LCBs. Phages isolated in this study, including Phage 30, Phage 15, and

Phage 18, comprised 16, 17, and 17 LCBs, respectively. All 13 cluster F phages included in

the comparison comprised all five LCBs. The comparison confirmed the presence of

homologue regions between the phages within the clusters but also highlighted that re -

arrangement and/or gain/loss of LCBs must have occurred at some point during the evolution

of the phages. The region encoding the terminase large subunit and portal protein were present

in a conserved region all genomes in all eight comparisons.


82

T

ab

le 2

| O

verv

iew

of

co

mpara

tiv

e g

en

om

ics

an

aly

sis

Clu

ster

# o

f

ph

ag

es

Ph

age

nam

es

# o

f

LC

Bs

# c

ore

gen

es

Core

gen

om

e

#

acc

esso

ry

gen

es

Acc

esso

ry

gen

om

e

Pan

gen

om

e

gen

es

A1*

1

7

Phag

e 77, 53, 75, 63, 54, 72, 64, 71, 47, 4

8, 59, 6

8, 65, 8,

Hen

u8, G

29

-2, an

d p

Sf-

1

16

2

8

22%

98

78%

1

26

A2*

7

Phag

e 80, 28, 56_1,7

6, S

wan

01, S

EC

phi2

7 a

nd E

coS_95

7

46

40%

69

60%

1

15

A3*

4

Phag

es 6

9, 52, 56_2, an

d J

ahat

_M

G145

4

62

73%

23

27%

8

5

B

6

Phag

e 73, 70, 17, 58, 74, an

d C

EB

_E

C3a

6

27

33%

54

67

8

1

C

10

Phag

e 61, T

5, E

AS

G3, H

AS

G4, A

KF

V33, O

SY

SP

,

phiL

LS

, S

P15, F

FH

1, an

d H

dH

2

6-1

0

86

37%

147

63%

2

33

D

19

Phag

e 60, 78, 62, 66, 79, E

C6, V

paE

1, X

TG

1, K

hF

1,

KhF

2, K

hF

3, H

Y02, R

o111lw

, O

157_1,

O157_12, W

V8,

O157_11, A

lf5, an

d A

YO

145A

14-1

7

72

39%

115

61%

1

87

E

18

Phag

e 30, 15, 18, H

X01, K

AW

3E

185, W

FbE

185, G

53,

AP

EC

c01, M

M02, H

P3, A

TK

47, A

TK

48, O

157_ 3

,

O157_ 6

, S

T0, JS

09, G

2285, an

d G

2469

16-1

7

184

58%

133

42%

3

17

F

13

Phag

e 55, 11, 10,

G2540-3

, G

29, G

4500, D

5505, G

9062,

CF

2, Y

UE

EL

01, fF

iEco

06,

AC

G_C

40, an

d O

E55

05

5

1

97

59%

136

41%

3

33

*S

ub

clu

ster in

stea

d o

f clu

ster

. Lo

cal c

ollin

ear b

locks

(LC

B) =

ind

ica

tin

g h

om

olo

go

us

DN

A regio

ns

sha

red

by tw

o o

r m

ore

gen

om

es

wit

ho

ut se

quence

rea

rrangem

ents

.


83

Discussion

In this study, 38 coliphages were isolated from poultry faecal material, sequenced and

characterised. The high number of coliphage genomes included in the analysis allowed for a

better understanding of both coliphage diversity in poultry and the global coliphage diversity

within the Siphoviridae and Myoviridae families. However, one should be aware of the possible

biases as the coliphages were isolated using two E. coli K12-derived host strains only and, as

such, cannot be seen as the complete coliphage diversity. Both host strains are mutated for the

FhuA (previously called TonA) gene, which is used as receptor for some phages, including

Siphoviridae coliphages T1 and T5 [40]. However, phage adsorption is not always restricted to

one receptor. Accordingly, we were able to isolate a T5-like phage using the selected host

strains. Also, as none of the host strains has the F pili, phages utilising a F pili encoded receptor

for absorption, such as Inoviridae phages, will most likely not be isolated [40]. The reasons

why we could not isolate coliphages from the Podoviridae family remains obscure, however,

this is in accordance with the finding of Korf et al. (2019), who also did not isolate any

Podoviridae coliphages from poultry while they could isolate them from sewage and surface

water. On the other hand, other studies have successfully isolated Podoviridae phages from

poultry faeces [41–44]. Several factors might play a role in the type of phage being isolated,

including culture and isolation method, host strain, and isolation source. In Podoviridae studies

mentioned above, phages were isolated using a single or multiple E. coli host-strains and the

DLA method was similar to this study. However, a notable difference is the type of host-strain

used. In our study, laboratory strains were used whereas the Podoviridae study host-strains

only included E. coli strains isolated from poultry. As our 11 Myoviridae phages were isolated

from samples from 11 different farms, no correlation between phage isolated and geographical

location (poultry farm) was found (data not shown). In accordance with the findings of Olsen

et al. (2020), the most prevalent genus of the Myoviridae and Siphoviridae phages isolated in

our study was the Felixounavirus (45.5%) and the Hanrivervirus (33.3%), respectively. Both

studies used E. coli K-12 derived laboratory strains as host strain for phage isolation. While in

a collection of 50 coliphages isolated from surface water, manure, sewage, or animal faeces,

29 different E. coli host strains were used [15]. No Felixounavirus phages could be isolated

and most Myoviridae phages belonged to the Tequatrovirus genus, followed by the Mosigvirus

genus. Those two genera were also found in our study.

Currently, the polyphasic approach is the most commonly used for bacterial classification [45,

46], and a similar approach combining multiple methods is recommended when working with


84

phages [12]. However, studying phage taxonomy has proven challenging since no universal

conserved marker gene, as the 16S rRNA gene used for bacteria, exists throughout all phage

families. Several semi-conserved family-specific marker genes have been proposed as

candidates to support taxonomical classification of tailed phages, including DNA packaging

and head assembly genes [1, 47, 48]. Accordingly, in this study we used the terminase large

subunit, portal protein and exonucleases as markers, and were able to determine the family and

subfamily to all coliphages isolated in this study in accordance with the TEM-based and whole

genome-based classification, respectively. Furthermore, the analysis based on the terminase

large subunit and portal protein resulted in a clear distinction between the two Tevenvirinae

clusters, indicating different genera (Mosigvirus and Tequatrovirus). Thus, single gene-based

analysis provides good initial indication in which taxonomic cluster the phages belong to.

However, a single gene does not provide a global view of the structural organisation of the

phage nor accounts for genomic rearrangements, mutations, and mosaicism. Moreover, in

accordance with findings from previous studies the selected gene was not always detected in

the phage genomes [47], thus excluding these phages from classification. When using single

genes, one should be aware of the possibility of multiple distinct variants of the same gene

within one genome. Several of our phages encoded up to three different exonuclease genes.

Depending on which gene variant used for analysis, the risk of “false” cluster formation and

distance, and hereby a faulty classification at subfamily or genus level was present. Thus, for

more comprehensive phage taxonomy, including genus classification, the single gene analysis

should be accompanied by whole genome-based analysis as well as functional gene studies.

When investigating the evolutionary relationship between phages studies have shown the

advantage of combining different proteomic and comparative genomic approaches, including

WGS data and well-characterised reference dataset, which take into account the effect of

horizontal gene transfer (HGT) and recombination events on the phage genome evolution [7,

49].

In this study, genome-based phylogenetic and taxonomic analysis were performed in

combination with traditional morphological examination of the phage using TEM. Through the

genome-based analysis we identified a potential new Siphoviridae genus. The three

unclassified A3 subcluster from this study clustered together with the Jahat_MG145 reference

phage, which was a singleton [16]. Thus, we propose that this group of phages represents a

new unclassified genus with currently four phages, including Phage 52, Phage 56_2 Phage 69,

and Jahat_MG145.


85

We aimed to expand our knowledge on the coliphage diversity, and observed great diversity

among these phages, while they were isolated from a similar ecosystem. The diversity was

characterised by a great span in genome size (44.3 kb to 173.1 kb) and G+C content range

(35.5-46.4%), as well as cluster-specific characteristics of the six phage clusters, A-F. Cluster

B phages had the smallest genomes and lowest number of CDSs followed by phages from

cluster A, D, C, and E/F. Similar to findings from other studies[16, 50], lower genome size

seemed to be correlated with an increase in G+C content. The largest variation in genome size

and number of CDSs were observed for the group of Myoviridae phages, whereas the largest

variation in G+C content (7.2%) was overserved for the group of Siphoviridae phages. Notably,

the Tequintavirus Phage 61 showed to be more similar to the group of Myoviridae compared

to the other Siphoviridae phages based on the above-mentioned characteristics. When omitting

Phage 61, G+C content variation for the Siphoviridae phages was only 2.9%. Phage G+C

content has been shown to be correlated with the G+C content of the phage host [50, 51].

Accordingly, differences in G+C content observed for the coliphages might reflect phage-host

interactions and co-evolution with past and current host(s). Gene content, including number of

core genes appeared to be associated with the cluster. Moreover, number of exonucleases

encoded by the phage appeared to be cluster associated as well, as only phages from cluster C

and cluster F encoded multiple exonucleases. Encoding multiple exonucleases could be a result

of adaptive evolution conferring fitness advantage over other phages. However, it could just as

well reflect some of the challenges to accurate phage genome annota tion, including false

negatives (undetected genes) and incorrect functional annotation [52, 53]. Gene content

variation has been shown to be related to recombination events resulting in acquisition or loss

of gene(s) [54]. Through our comparative genomics analysis of related phages, LCBs with

modules with varying level of gene conservation were identified and highlighted the different

levels of heterogeneity between different phage clusters. Repeat regions were observed in

several of the phage genomes and resulted in variation of LCBs. The presence of these regions

should be considered when assessing the gene content variation, as these regions are shown to

be prone to genome assembly mistakes, and as such, might represent false level of variation

[55]. LCBs modules are also called for mosaic sections and the two terms have been used

interchangeably throughout history, referring to exchangeable genomic segments between two

or more phages in the population [13]. The genome comparison showed the mosaic nature of

the phage genomes, with modules with high level of conservation interspersed with low-

similarity sections. These sections could be acquired though HGT from other phages, which is

thought to happen when phages are found in the same host. Most often this happens through


86

phage co-infection or single-phage infection of a host that carries one or more prophages [56,

57]. Moreover, phages have been shown to acquire genes from their host [58]. The comparative

genomics approach hereby underlines the continuous evolution of phage genomes as well as

the great phage diversity.

In conclusion, this study has identified a potential new coliphage genus and several new

species and provides insight not only to the coliphage diversity of the intestine of poultry but

the global coliphage diversity as well. Moreover, classification of phages isolated in this study

brings us one step closer to a more refined taxonomic understanding of coliphages. Our

comparative genomic analysis showed different levels of heterogeneity between different

phage clusters and highlighted the mosaic nature of the phage genomes as well as the

continuous evolution of phages in a single environment source. However, to fully understand

the complexity and underlying mechanisms of the phage diversity further studies are needed.


87

References






virology-100114-054952.









[6] Ackermann, H. W. Phage Classification and Characterization. In Bacteriophages: Methods and

Protocols, Volume 1: Isolation, Characterization, and Interactions; Clokie, M. R. J., Kropinski, A.

M., Eds.; Methods in Molecular Biology; Humana Press: Totowa, NJ, 2009; Vol. 501, pp 127 –140.

https://doi.org/10.1007/978-1-60327-164-6.

[7] Rohwer, F.; Edwards, R. The Phage Proteomic Tree: A Genome-Based Taxonomy for Phage. J.


[8] Lefkowitz, E. J.; Dempsey, D. M.; Hendrickson, R. C.; Orton, R. J.; Siddell, S. G.; Smith, D. B.

Virus Taxonomy: The Database of the International Committee on Taxonomy of Viruses (ICTV).

Nucleic Acids Res., 2018, 46 (D1), D708–D717. https://doi.org/10.1093/nar/gkx932.

[9] Meier-Kolthoff, J. P.; Göker, M. VICTOR: Genome-Based Phylogeny and Classification of

Prokaryotic Viruses. Bioinformatics, 2017, 33 (21), 3396–3404.

https://doi.org/10.1093/bioinformatics/btx440.

[10] Adriaenssens, E. M.; Rodney Brister, J. How to Name and Classify Your Phage: An Informal Guide.

Viruses, 2017, 9 (4), 1–9. https://doi.org/10.3390/v9040070.

[11] International Committee on Taxonomy of Viruses (ICTV), A. M. K. Virus Taxonomy: Ninth Report

of the International Committee on Taxonomy of Viruses, 1st ed.; King, A. M. ., Adams, M. J.,

Carstens, E. B., Lefkowitz, E. J., Eds.; Elsevier, 2011.





[13] Casjens, S. R.; Thuman-Commike, P. A. Evolution of Mosaically Related Tailed Bacteriophage

Genomes Seen through the Lens of Phage P22 Virion Assembly. Virology, 2011, 411 (2), 393–415.


[14] Hatfull, G. F.; Hendrix, R. W. Bacteriophages and Their Genomes. Curr. Opin. Virol., 2011, 1 (4),

298–303. https://doi.org/10.1016/j.coviro.2011.06.009.

[15] Korf, I. H. E.; Meier-Kolthoff, J. P.; Adriaenssens, E. M.; Kropinski, A. M.; Nimtz, M.; Rohde, M.;

van Raaij, M. J.; Wittmann, J. Still Something to Discover: Novel Insights into Escherichia Coli

Phage Diversity and Taxonomy. Viruses, 2019, 11 (5), 1–29. https://doi.org/10.3390/v11050454.

[16] Olsen, N. S.; Kot, W.; Junco, L. M. F.; Hansen, L. H. Exploring the Remarkable Diversity of

Escherichia Coli Phages in the Danish Wastewater Environment, Including 91 Novel Phage Species.


88

bioRxiv, 2020.

[17] Adams, M. H. Bacteriophages; Interscience Publishers: New York, 1959.

[18] Bonilla, N.; Rojas, M. I.; Netto Flores Cruz, G.; Hung, S.-H.; Rohwer, F.; Barr, J. J. Phage on Tap–

a Quick and Efficient Protocol for the Preparation of Bacteriophage Laboratory Stocks. PeerJ, 2016.

https://doi.org/10.7717/peerj.2261.

[19] Appleyard, R. K. Segregation of New Lysogenic Types during Growth of a Doubly Lysogenic Strain

Derived from Escherichia Coli K12. Genetics, 1954, 39 (4), 440–452.

[20] Colson, C.; Glover, S. W.; Symonds, N.; Stacey, K. A. The Location of the Genes for Host -

Controlled Modification and Restriction in Escherichia Coli K-12. Genetics, 1965, 52 (5), 1043–

1050.

[21] Kropinski, A. M.; Mazzocco, A.; Waddell, T. E.; Lingohr, E.; Johnson, R. P. Enumeration of

Bacteriophages by Double Agar Overlay Plaque Assay. In Bacteriophages. Methods in Molecular

Biology; Clokie, M. R., Kropinski, A. M., Eds.; Humana Press, 2009; Vol. 501, pp 69 –76.

https://doi.org/10.1007/978-1-60327-164-6_7.

[22] Oliveira, A.; Sillankorva, S.; Quinta, R.; Henriques, A.; Sereno, R.; Azeredo, J. Isolation and

Characterization of Bacteriophages for Avian Pathogenic E. Coli Strains. J. Appl. Microbiol., 2009,

106 (6), 1919–1927. https://doi.org/10.1111/j.1365-2672.2009.04145.x.

[23] Jakočiūnė, D.; Moodley, A. A Rapid Bacteriophage DNA Extraction Method. Methods Protoc.,

2018, 1 (3), 27. https://doi.org/10.3390/mps1030027.

[24] Kot, W.; Vogensen, F. K.; Sørensen, S. J.; Hansen, L. H. DPS - A Rapid Method for Genome

Sequencing of DNA-Containing Bacteriophages Directly from a Single Plaque. J. Virol. Methods,

2014, 196, 152–156. https://doi.org/10.1016/j.jviromet.2013.10.040.

[25] Altschul, S. F.; Gish, W.; Miller, W.; Myers, E. W.; Lipman, D. J. Basic Local Alignment Search

Tool. J. Mol. Biol., 1990, 215 (3), 403–410. https://doi.org/10.1016/S0022-2836(05)80360-2.

[26] Arndt, D.; Grant, J. R.; Marcu, A.; Sajed, T.; Pon, A.; Liang, Y.; Wishart, D. S. PHASTER: A Better,

Faster Version of the PHAST Phage Search Tool. Nucleic Acids Res., 2016, 44, W16–W21.

https://doi.org/10.1093/nar/gkw387.

[27] Zankari, E.; Hasman, H.; Cosentino, S.; Vestergaard, M.; Rasmussen, S.; Lund, O.; Aarestrup, F.

M.; Larsen, M. V. Identification of Acquired Antimicrobial Resistance Genes. J. Antimicrob.

Chemother., 2012, 67 (11), 2640–2644. https://doi.org/10.1093/jac/dks261.

[28] Joensen, K. G.; Scheutz, F.; Lund, O.; Hasman, H.; Kaas, R. S.; Nielsen, E. M.; Aarestrup, F. M.

Real-Time Whole-Genome Sequencing for Routine Typing, Surveillance, and Outbreak Detection

of Verotoxigenic Escherichia Coli. J. Clin. Microbiol., 2014, 52 (5), 1501–1510.

https://doi.org/10.1128/JCM.03617-13.

[29] Overbeek, R.; Olson, R.; Pusch, G. D.; Olsen, G. J.; Davis, J. J.; Disz, T.; Edwards, R. A.; Gerdes,

S.; Parrello, B.; Shukla, M.; et al. The SEED and the Rapid Annotation of Microbial Genomes Using

Subsystems Technology (RAST). Nucleic Acids Res., 2014, 42 (D1), 206–214.

https://doi.org/10.1093/nar/gkt1226.

[30] R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for

Statistical Computing: Vienna, Austria 2020.

[31] Page, A. J.; Cummins, C. A.; Hunt, M.; Wong, V. K.; Reuter, S.; Holden, M. T. G.; Fookes, M.;

Falush, D.; Keane, J. A.; Parkhill, J. Roary: Rapid Large-Scale Prokaryote Pan Genome Analysis.

Bioinformatics, 2015, 31 (22), 3691–3693. https://doi.org/10.1093/bioinformatics/btv421.

[32] Seemann, T. Prokka: Rapid Prokaryotic Genome Annotation. Bioinformatics, 2014, 30 (14), 2068–

2069. https://doi.org/10.1093/bioinformatics/btu153.

[33] Hyatt, D.; Chen, G.; Locascio, P. F.; Land, M. L.; Larimer, F. W.; Hauser, L. J. Prodigal: Prokaryotic


89

Gene Recognition and Translation Initiation Site Identification. BMC Bioinformatics, 2010, 11 (119).

https://doi.org/10.1186/1471-2105-11-119.

[34] Larkin, M. A.; Blackshields, G.; Brown, N. P.; Chenna, R.; Mcgettigan, P. A.; McWilliam, H.;

Valentin, F.; Wallace, I. M.; Wilm, A.; Lopez, R.; et al. Clustal W and Clustal X Version 2.0.

Bioinformatics, 2007, 23 (21), 2947–2948. https://doi.org/10.1093/bioinformatics/btm404.

[35] Kumar, S.; Stecher, G.; Li, M.; Knyaz, C.; Tamura, K. MEGA X: Molecular Evolutionary Genetics

Analysis across Computing Platforms. Mol. Biol. Evol., 2018, 35 (6), 1547–1549.

https://doi.org/10.1093/molbev/msy096.

[36] Paradis, E.; Schliep, K. Ape 5.0: An Environment for Modern Phylogenetics and Evolutionary

Analyses in R. Bioinformatics, 2019, 35 (3), 526–528.

https://doi.org/10.1093/bioinformatics/bty633.

[37] Schliep, K. P. Phangorn: Phylogenetic Analysis in R. Bioinformatics, 2011, 27 (4), 592–593.

https://doi.org/10.1093/bioinformatics/btq706.

[38] Darling, A. E.; Mau, B.; Perna, N. T. Progressivemauve: Multiple Genome Alignment with Gene

Gain, Loss and Rearrangement. PLoS One, 2010, 5 (6).

https://doi.org/10.1371/journal.pone.0011147.

[39] Besler, I.; Sazinas, P.; Harrison, C.; Gannon, L.; Redgwell, T.; Michniewski, S.; Hooton, S. P.;

Hobman, J. L.; Millard, A. Genome Sequence and Characterisation of Coliphage VB_Eco_SLUR29.

PHAGE Ther. Appl. Res., 2020, 1 (1), 38–44. https://doi.org/10.1089/phage.2019.0009.

[40] Bertozzi Silva, J.; Storms, Z.; Sauva geau, D. Host Receptors for Bacteriophage Adsorption. FEMS


[41] Mullally, J.; Stockdale, Stephen R. Smith, M. K.; Draper, L. A.; Ross, P. R.; Hill, C. Complete

Genome Sequence of Escherichia Coli Phage APC_JM3.2 Isolated from a Chicken Cecum. Genome

Announc., 2018, 6 (1), e01292-17. https://doi.org/10.1128/genomeA.01292-17.

[42] Li, Y.; Chen, M.; Tang, F.; Yao, H.; Lu, C.; Zhang, W. Complete Genome Sequence of the Novel

Lytic Avian Pathogenic Coliphage NJ01. J. Virol., 2012, 86 (24), 13874–13875.

https://doi.org/10.1128/jvi.02727-12.

[43] Chen, M.; Xu, J.; Yao, H.; Lu, C.; Zhang, W. Isolation, Genome Sequencing and Functional Analysis

of Two T7-like Coliphages of Avian Pathogenic Escherichia Coli. Gene, 2016, 582 (1), 47–58.

https://doi.org/10.1016/j.gene.2016.01.049.

[44] Lau, G. L.; Sieo, C. C.; Tan, W. S.; Ho, Y. W. Characteristics of a Phage Effective for Colibacillosis

Control in Poultry. J. Sci. Food Agric., 2012, 92 (13), 2657–2663. https://doi.org/10.1002/jsfa.5683.

[45] Vandamme, P.; Pot, B.; Gillis, M.; De Vos, P.; Kersters, K.; Swings, J. Polyphasic Taxonomy, a

Consensus Approach to Bacterial Systematics. Microbiol. Rev., 1996, 60 (2), 407–438.

https://doi.org/10.1128/mmbr.60.2.407-438.1996.

[46] Prakash, O.; Verma, M.; Sharma, P.; Kumar, M.; Kumari, K.; Singh, A.; Kumari, H.; Jit, S.; Gupta,

S. K.; Khanna, M.; et al. Polyphasic Approach of Bacterial Classification - An Overview of Recent

Advances. Indian J. Microbiol., 2007, 47 (2), 98–108. https://doi.org/10.1007/s12088-007-0022-x.

[47] Lopes, A.; Tavares, P.; Petit, M. A.; Guérois, R.; Zinn-Justin, S. Automated Classification of Tailed

Bacteriophages According to Their Neck Organization. BMC Genomics, 2014, 15, 1027.

https://doi.org/10.1186/1471-2164-15-1027.

[48] Sullivan, M. B.; Coleman, M. L.; Quinlivan, V.; Rosenkrantz, J. E.; DeFrancesco, A. S.; Tan, G.; Fu,

R.; Lee, J. A.; Waterbury, J. B.; Bielawski, J. P.; et al. Portal Protein Diversity and Phage Ecology.


[49] Glazko, G.; Makarenkov, V.; Liu, J.; Mushegian, A. Evolutionary History of Bacteriophages with

Double-Stranded DNA Genomes. Biol. Direct, 2007, 2 (36), 1–14. https://doi.org/10.1186/1745-

6150-2-36.


90

[50] Almpanis, A.; Swain, M.; Gatherer, D.; McEwan, N. Correlation between Bacterial G+C Content,

Genome Size and the G+C Content of Associated Plasmids and Bacteriophages. Microb. Genomics,

2018, 4 (4), e000168. https://doi.org/10.1099/mgen.0.000168.

[51] Jacobs-Sera, D.; Marinelli, L. J.; Bowman, C.; Broussard, G. W.; Guerrero Bustamante, C.; Boyle,

M. M.; Petrova, Z. O.; Dedrick, R. M.; Pope, W. H.; Modlin, R. L.; et al. On the Nature of

Mycobacteriophage Diversity and Host Preference. Virology, 2012, 434 (2), 187–201.


[52] McNair, K.; Aziz, R. K.; Pusch, G. D.; Overbeek, R.; Dutilh, B. E.; Edwards, R. Phage Genome

Annotation Using the RAST Pipeline. In Bacteriophages. Methods in Molecular Biology; Humana

Press, New York, NY, 2018; Vol. 1681, pp 231–238. https://doi.org/10.1007/978-1-4939-7343-

9_17.

[53] Salisbury, A.; Tsourkas, P. K. A Method for Improving the Accuracy and Efficiency of

Bacteriophage Genome Annotation. Int. J. Mol. Sci., 2019, 20 (3391), 1–17.

https://doi.org/10.3390/ijms20143391.

[54] Kupczok, A.; Neve, H.; Huang, K. D.; Hoeppner, M. P.; Heller, K. J.; Franz, C. M. A. P.; Dagan, T.

Rates of Mutation and Recombination in Siphoviridae Phage Genome Evolution over Three

Decades. Mol. Biol. Evol., 2018, 35 (5), 1147–1159. https://doi.org/10.1093/molbev/msy027.

[55] Klumpp, J.; Fouts, D. E.; Sozhamannan, S. Next Generation Sequencing Technologies and the

Changing Landscape of Phage Genomics. Bacteriophage, 2012, 2 (3), 190–199.

https://doi.org/10.4161/bact.22111.

[56] Hendrix, R. W. Bacteriophage Genomics. Curr. Opin. Microbiol., 2003, 6 (5), 506–511.

https://doi.org/10.1016/j.mib.2003.09.004.

[57] Hendrix, R. W.; Smith, M. C. M.; Burns, R. N.; Ford, M. E.; Hatfull, G. F. Evolutionary

Relationships among Diverse Bacteriophages and Prophages: All the World’s a Phage. Proc. Natl.

Acad. Sci., 1999, 96 (5), 2192–2197. https://doi.org/10.1073/pnas.96.5.2192.

[58] Pedulla, M. L.; Ford, M. E.; Houtz, J. M.; Karthikeyan, T.; Wadsworth, C.; Lewis, J. A.; Jacobs-

Sera, D.; Falbo, J.; Gross, J.; Pannunzio, N. R.; et al. Origins of Highly Mosaic Mycobacteriophage

Genomes. Cell, 2003, 113 (2), 171–182. https://doi.org/10.1016/S0092-8674(03)00233-2.


91

Supplementary Script Kmer.py

#!/usr/bin/env python3

import Utils

import argparse

import sys

import math

import collections

#import timeit

parser=argparse.ArgumentParser(description="Creates a kmer profile from a fastq or fasta file")

parser.add_argument('seq_files',type=str,nargs='+')

parser.add_argument('-l','--length',type=int,default=10)

parser.add_argument('-e','--euclidian',action="store_true")

args=parser.parse_args()

kmerlength=args.length

def kmer_fq(fil):

kmers=dict()

fqs=Utils.FqStream(fh=fil)

for fq in fqs:

for i in range(len(fq)-kmerlength):

try:

kmers[fq.sseq()[i:i+kmerlength]]+=1.0

except KeyError:

kmers[fq.sseq()[i:i+kmerlength]]=1.0

unfounded=list()

for (k,v) in kmers.items():

try:

if kmers[k[::-1].translate(Utils.tr)] + v <3:

unfounded.append(k)

except KeyError:

unfounded.append(k)

for k in unfounded:

del kmers[k]

if args.euclidian:

normalize(kmers)

return kmers

def kmer_fasta(fil):

kmers=dict()

try:

fasta="".join([line.decode().strip() if line.decode()[0]!=">" else "" for line in fil])

except AttributeError:

fasta="".join([line.strip() if line[0]!=">" else "" for line in fil])

for i in range(len(fasta)-kmerlength):

try:

kmers[fasta[i:i+kmerlength]]+=1.0

except KeyError:

kmers[fasta[i:i+kmerlength]]=1.0

if args.euclidian:

normalize(kmers)

return kmers


92

def normalize(kmers):

s=0.0

for val in kmers.values():

s+=val*val

s=math.sqrt(s)

for k in kmers.keys():

kmers[k]/=s

return kmers

def dist2(kmer1,kmer2):

dist=0.0

(L1,L2)=(sum(kmer1.values()),sum(kmer2.values()))

for (k,v) in kmer1.items():

try:

dist+=min(v,kmer2[k])

except KeyError:

pass

return dist/(min(L1,L2))

def dist(kmer1,kmer2):

dist=0.0

ignore=set()

for (k,v) in kmer1.items():

try:

d=v-kmer2[k]

dist+=d*d

ignore.add(k)

except KeyError:

dist+=v*v

remainder=set(kmer2.keys()).difference(ignore)

for k in remainder:

d=kmer2[k]

dist+=d*d

return math.sqrt(dist)

kmers=list()

print("Reading data...",file=sys.stderr)

for filename in args.seq_files:

fil=Utils.gzopen(filename)

line=next(fil)

try:

line=line.decode()

except AttributeError:

pass

if line.startswith(">"):

filetype="Fasta"

kmers.append(kmer_fasta(fil))

elif line.startswith("@"):

filetype="Fastq"

fil.seek(0)

kmers.append(kmer_fq(fil))

else:

print("Unrecognized file format:\n{}".format(line),file=sys.stdout)

fil.close()

print("Done",file=sys.stderr)

for i in range(len(kmers)):

print(args.seq_files[i],end="\t")


93

for j in range(i):

print("\t",end="")

for j in range(i+1,len(kmers)):

print("\t{}".format(dist(kmers[i],kmers[j]) if args.euclidian else dist2(kmers[i],kmers[j])),end="")

print("")


94

Su

pp

lem

en

tary

Ta

ble

S1

| C

hara

cte

rist

ics

of

refe

ren

ce g

en

om

es m

en

tio

ned in

th

is s

tud

y

Ph

ag

e n

am

e P

ha

ge

Gen

om

e

size (b

p)

G+

C%

#

CD

Ss

Ph

ag

e

fam

ily

Ph

ag

e

sub

fam

ily

Ph

ag

e

clu

ster

Ph

ag

e g

en

us

Access

ion

nu

mb

er

pS

f-1

Shig

ella

phag

e pS

f-1

51821

44.0

93

Sip

hovi

ridae

Tunavi

rinae

A1

Hanri

verv

irus

KC

710998.1

YU

EE

L01

E

scher

ichia

phag

e Y

UE

EL

01

169621

35.4

268

Myo

viri

dae

Tev

envi

rinae

F

Teq

uatr

ovi

rus

KY

290975.2

fFiE

co0

6

Esc

her

ich

ia p

hag

e v

B_

Eco

M-f

FiE

co06

167076

35.5

267

M

yovi

ridae

Tev

envi

rinae

F

Teq

uatr

ovi

rus

MG

781190.1

CE

B_

EC

3a

Esc

her

ich

ia p

hag

e v

B_

Eco

s_C

EB

_E

C3

a 44234

44.2

72

Sip

hovi

ridae

Tu

navi

rinae

B

Rtp

viru

s K

Y398841.1

ST

0

Esc

her

ichia

phag

e S

T0

170496

37.7

270

Myo

viri

dae

Tev

envi

rinae

E

Mosi

gvi

rus

MF

044457.1

O157_ 3

E

scher

ichia

coli O

157

typin

g p

hag

e 3

168733

37.3

268

Myo

viri

dae

Tev

envi

rinae

E

Mosi

gvi

rus

KP

869101.1

Sw

an0

1

Esc

her

ich

ia p

hag

e v

B_

Eco

_sw

an0

1

50865

44.7

85

Sip

hovi

ridae

Tu

navi

rinae

A2

Uncl

assi

fied

L

T841304.1

RB

14

E

nte

rob

acte

ria

ph

age

RB

14

165429

35.3

264

M

yovi

ridae

Tev

envi

rinae

F

Teq

uatr

ovi

rus

FJ8

39692.1

Alf

5

Esc

her

ichia

phag

e vB

_E

coM

_A

lf5

87662

39.0

130

Myo

viri

dae

Ounavi

rinae

C

Fel

ixounavi

rus

KX

377933.1

OS

YS

P

Esc

her

ichia

phag

e O

SY

SP

110901

39.2

161

Sip

hovi

ridae

N/A

D

T

equin

tavi

rus*

M

F402939.1

VpaE

1

Esc

her

ichia

phag

e vB

_E

coM

-VpaE

1

88403

38.9

126

Myo

viri

dae

Ounavi

rinae

C

Fel

ixounavi

rus

KM

657822.1

AY

O1

45

A

Esc

her

ich

ia p

hag

e v

B_

Eco

M_

AY

O14

5A

87372

39.0

128

M

yovi

ridae

Ou

navi

rinae

C

Fel

ixounavi

rus

KR

014248.1

SU

10

E

sch

eric

hia

Ph

age

vB

_E

coP

_S

U10

77327

42.1

127

P

odovi

ridae

N/A

-

Kura

viru

s*

KM

044272.1

Eco

S_95

Esc

her

ichia

phag

e vB

_E

coS

-95

50910

44.8

87

Sip

hovi

ridae

Uncl

assi

fied

A

2

Uncl

assi

fied

M

F564201.1

Jahat

_M

G145

Esc

her

ichia

phag

e Ja

hat

_M

G145

50984

45.7

87

Sip

hovi

ridae

Tunavi

rinae

A3

Uncl

assi

fied

M

K552105.1

Lam

bd

a E

nte

rob

acte

ria

ph

age

lam

bd

a

48502

49.9

71

Sip

hovi

ridae

Tu

navi

rinae

- L

am

bdavi

rus

J02459.1

T1

E

nte

rob

acte

ria

ph

age

T1

48836

45.6

76

Sip

hovi

ridae

Tu

navi

rinae

B

Tunavi

rus

AY

216660.1

T4

E

nte

rob

acte

ria

ph

age

T4

168903

35.3

266

M

yovi

ridae

Tev

envi

rinae

F

Teq

uatr

ovi

rus

AF

158101.6

T5

B

acte

riophag

e T

5

121750

39.3

172

Sip

hovi

ridae

N/A

D

T

equin

tavi

rus*

A

Y543070.1

T7

E

nte

robac

teri

a phag

e T

7

39937

48.4

52

Podovi

ridae

Auto

gra

phiv

irin

ae

- T

esep

tim

avi

rus

NC

_001604.1

P1

E

sch

eric

hia

vir

us

P1

66750

47.6

92

M

yovi

ridae

N/A

-

Punavi

rus*

M

H445381.1

P2

E

nte

rob

acte

ria

ph

age

P2

31200

52.6

49

M

yovi

ridae

Ped

uovi

rinae

- P

eduovi

rus

NC

_041848.1

P22

E

nte

robac

teri

a phag

e P

22

41724

47.1

68

Podovi

ridae

N/A

-

Led

erber

gvi

rus*

N

C_002371.2

Mu

E

nte

robac

teri

a phag

e M

u

36717

52.1

54

Myo

viri

dae

N/A

-

Muvi

rus*

N

C_000929.1

Hen

u8**

E

scher

ichia

phag

e H

enu8

49890

44.2

85

Sip

hovi

ridae

Tunavi

rinae

A1

Hanri

verv

irus

MN

055691.1

G2

9_

2*

*

Esc

her

ich

ia p

hag

e v

B_

Eco

S_G

29-2

51739

44.0

89

Sip

hovi

ridae

Tu

navi

rinae

A1

Hanri

verv

irus

MK

373798.1


95

S

up

ple

men

tary

Ta

ble

S1

| C

on

tin

ued

Ph

ag

e n

am

e P

ha

ge

Gen

om

e

size (b

p)

G+

C%

#

CD

Ss

Ph

ag

e

fam

ily

Ph

ag

e

sub

fam

ily

Ph

ag

e

clu

ster

P

ha

ge g

en

us

Access

ion

nu

mb

er

SE

C_phi2

7**

E

scher

ichia

phag

e S

EC

ph

i27

51811

44.7

87

Sip

hovi

ridae

Tuna

viri

nae

A2

Uncl

assi

fied

L

T961732.1

AC

G_M

12

Ente

robac

teri

a phag

e vB

_E

coS

_A

CG

-M12

46054

44.0

77

Sip

hovi

ridae

Tuna

viri

nae

B

Rtp

viru

s N

C_019404.1

Go

lest

an

Esc

her

ich

ia p

hag

e V

B_

Eco

S-G

ole

stan

44829

50.6

78

Sip

hovi

ridae

Guer

nse

yvir

ina

e (B

) K

agunavi

rus

MG

099933.1

MM

01

Esc

her

ich

ia p

hag

e v

B_

Eco

S_

MM

01

43157

43.8

69

Sip

hovi

ridae

Tun

avi

rina

e (B

) R

ogunavi

rus

MK

373793.1

VA

H1**

Esc

her

ichia

phag

e v

B_E

coS

_V

AH

1

124537

38.6

206

Sip

hovi

ridae

N/A

C

T

equin

tavi

rus*

M

K373792.1

EA

SG

3**

Esc

her

ichia

phag

e v

B_E

coS

_E

AS

G3

120715

39.0

181

Sip

hovi

ridae

N/A

C

T

equin

tavi

rus*

M

K373799.1

HA

SG

4*

*

Esc

her

ich

ia p

hag

e v

B_

Eco

S_

HA

SG

4

120603

39.0

182

Sip

hovi

ridae

N/A

C

T

equin

tavi

rus*

M

K373797.1

AK

FV

33

E

sch

eric

hia

ph

age

bV

_E

coS

_A

KF

V33

108853

38.9

160

Sip

hovi

ridae

N/A

C

T

equin

tavi

rus*

N

C_017969.1

ph

iLL

S*

*

Esc

her

ich

ia _

ph

age_

ph

iLL

S

107263

39.0

156

Sip

hovi

ridae

N/A

C

T

equin

tavi

rus*

K

Y677846.1

SP

15**

E

scher

ichia

_ph

age_

SP

15_

DN

A

110964

39.1

160

Sip

hovi

ridae

N/A

C

T

equin

tavi

rus*

A

P019559.1

FF

H1

E

scher

ichia

phag

e v

B_E

coS

_F

FH

1

108483

39.2

155

Sip

hovi

ridae

N/A

C

T

equin

tavi

rus*

N

C_024139.1

Hd

H2

**

E

sch

eric

hia

ph

age

vB

_E

coS

_H

dH

2

120120

39.3

174

Sip

hovi

ridae

N/A

C

T

equin

tavi

rus*

M

K373796.1

H8

Esc

her

ich

ia v

iru

s H

8

104373

38.8

149

Sip

hovi

ridae

N/A

C

T

equin

tavi

rus*

N

C_042307.1

DT

57C

E

nte

robac

teri

a_phag

e_D

T57C

108065

39.7

149

Sip

hovi

ridae

N/A

C

T

equin

tavi

rus*

N

C_027356.1

Gost

ya9

**

E

scher

ichia

_ph

age_

Gost

ya9

101665

39.4

137

Sip

hovi

ridae

N/A

C

T

equin

tavi

rus*

M

H203051.1

EC

6

Esc

her

ichia

phag

e E

C6

86231

38.9

137

Myo

viri

dae

Ouna

viri

nae

D

Fel

ixounavi

rus

NC

_027369.1

JH2

E

sch

eric

hia

ph

age

JH2

87712

38.8

132

M

yovi

ridae

Oun

avi

rina

e D

F

elix

ounavi

rus

NC

_029023.1

XT

G1

**

E

nte

rob

acte

ria

ph

age

XT

G1

89635

38.9

130

M

yovi

ridae

Oun

avi

rina

e D

F

elix

ounavi

rus

KT

184316.1

KhF

1**

Ente

robac

teri

a phag

e K

hF

1

88356

38.8

133

Myo

viri

dae

Ouna

viri

nae

D

Fel

ixounavi

rus

KT

184313.1

KhF

2**

Ente

robac

teri

a phag

e K

hF

2

88309

38.8

131

Myo

viri

dae

Ouna

viri

nae

D

Fel

ixounavi

rus

KT

184314.1

Kh

F3

**

E

nte

rob

acte

ria

ph

age

Kh

F3

88016

38.9

130

M

yovi

ridae

Oun

avi

rina

e D

F

elix

ounavi

rus

KT

184315.1

HY

02

Esc

her

ich

ia p

hag

e H

Y0

2

86252

38.9

124

M

yovi

ridae

Oun

avi

rina

e D

F

elix

ounavi

rus

NC

_028872.1

Ro111lw

**

E

scher

ichia

phag

e v

B_E

coM

-Ro111lw

86950

38.8

126

Myo

viri

dae

Ouna

viri

nae

D

Fel

ixounavi

rus

MH

571750.1

O157_ 1

**

Esc

her

ichia

coli

O157

typin

g p

hag

e 1

88531

38.8

135

Myo

viri

dae

Ouna

viri

nae

D

Fel

ixounavi

rus

NC

_041979.1

O157_ 1

1**

E

scher

ichia

coli

O157

typin

g p

hag

e 1

1 88771

38.9

134

Myo

viri

dae

Ouna

viri

nae

D

Fel

ixounavi

rus

KP

869109.1

O1

57

_ 1

2*

*

Esc

her

ich

ia c

oli

O1

57

ty

pin

g p

hag

e 1

2

88632

38.9

132

M

yovi

ridae

Oun

avi

rina

e D

F

elix

ounavi

rus

KP

869110.1


96

Su

pp

lem

en

tary

Ta

ble

S1

| C

on

tin

ued

Ph

ag

e n

am

e

Ph

ag

e

Gen

om

e

size (b

p)

G+

C%

#

CD

Ss

Ph

ag

e

fam

ily

Ph

ag

e

sub

fam

ily

Ph

ag

e

clu

ster

P

ha

ge g

en

us

Access

ion

nu

mb

er

WV

8

Ente

robac

teri

a phag

e W

V8

8

84

87

38.9

134

Myo

viri

dae

Ounavi

rinae

D

Fel

ixounavi

rus

NC

_012749.1

HX

01

Ente

robac

teri

a phag

e H

X01

1

69

15

8

37.6

268

Myo

viri

dae

Tev

envi

rinae

E

Mosi

gvi

rus

NC

_018855.1

KA

W3E

185**

Esc

her

ich

ia p

hag

e

vB

_E

coM

_K

AW

3E

185

1

70

18

7

37.6

270

Myo

viri

dae

Tev

envi

rinae

E

Mosi

gvi

rus

MK

373782.1

WF

bE

185**

E

scher

ichia

phag

e vB

_E

coM

_W

FbE

185

1

70

42

9

37.6

272

Myo

viri

dae

Tev

envi

rinae

E

Mosi

gvi

rus

MK

373778.1

G5

3*

*

Esc

her

ich

ia p

hag

e v

B_

Eco

M_G

53

1

67

83

4

37.8

268

M

yovi

ridae

Tev

envi

rinae

E

Mosi

gvi

rus

MK

327943.1

AP

EC

c01

E

sch

eric

hia

ph

age

AP

EC

c01

1

68

77

1

37.7

267

M

yovi

ridae

Tev

envi

rinae

E

Mosi

gvi

rus

NC

_029091.1

MM

02

**

E

sch

eric

hia

ph

age

vB

_E

coM

_M

M02

1

69

20

1

37.6

269

M

yovi

ridae

Tev

envi

rinae

E

Mosi

gvi

rus

MK

373784.1

HP

3

Esc

her

ichia

phag

e H

P3

1

70

25

4

37.6

268

Myo

viri

dae

Tev

envi

rinae

E

Mosi

gvi

rus

NC

_041920.1

AT

K47**

E

nte

robac

teri

a phag

e A

TK

47

1

70

02

0

37.6

270

Myo

viri

dae

Tev

envi

rinae

E

Mosi

gvi

rus

KT

184309.1

AT

K4

8*

*

En

tero

bac

teri

a p

hag

e A

TK

48

1

69

72

9

37.6

270

M

yovi

ridae

Tev

envi

rinae

E

Mosi

gvi

rus

KT

184310.1

O1

57

_ 6

E

sch

eric

hia

co

li O

15

7 ty

pin

g p

hag

e 6

1

60

57

0

37.6

249

M

yovi

ridae

Tev

envi

rinae

E

Mosi

gvi

rus

NC

_041864.1

JS09

E

scher

ichia

phag

e vB

_E

coM

_JS

09

1

69

14

8

37.6

271

Myo

viri

dae

Tev

envi

rinae

E

Mosi

gvi

rus

NC

_024124.2

G2285**

Esc

her

ichia

phag

e vB

_E

coM

_G

2285

1

66

67

5

37.5

262

Myo

viri

dae

Tev

envi

rinae

E

Mosi

gvi

rus

MK

327933.1

G2469**

Esc

her

ichia

phag

e vB

_E

coM

_G

2469

1

70

45

2

37.6

271

Myo

viri

dae

Tev

envi

rinae

E

Mosi

gvi

rus

MK

327934.1

G2

54

0_

3*

*

Esc

her

ich

ia p

hag

e v

B_

Eco

M_G

2540

-3

16

86

54

35.3

271

M

yovi

ridae

Tev

envi

rinae

F

Teq

uatr

ovi

rus

MK

327944.1

G2

9*

*

Esc

her

ich

ia p

hag

e v

B_

Eco

M_G

29

1

68

24

1

35.3

269

M

yovi

ridae

Tev

envi

rinae

F

Teq

uatr

ovi

rus

MK

327940.1

G4500**

Esc

her

ichia

phag

e vB

_E

coM

_G

4500

1

68

36

3

35.3

268

Myo

viri

dae

Tev

envi

rinae

F

Teq

uatr

ovi

rus

MK

327945.1

D5505**

Esc

her

ichia

phag

e D

5505

1

68

04

9

35.4

269

Myo

viri

dae

Tev

envi

rinae

F

Teq

uatr

ovi

rus

MK

327929.1

G9

06

2*

*

Esc

her

ich

ia p

hag

e v

B_

Eco

M_G

9062

1

68

67

0

35.3

268

M

yovi

ridae

Tev

envi

rinae

F

Teq

uatr

ovi

rus

MK

373779.1

CF

2

Esc

her

ich

ia p

hag

e C

F2

1

68

18

8

35.4

264

M

yovi

ridae

Tev

envi

rinae

F

Teq

uatr

ovi

rus

NC

_041919.1

AC

G_

C4

0*

*

Ente

robac

teri

a phag

e vB

_E

coM

_A

CG

-

C40

16

73

96

35.2

273

M

yovi

ridae

Tev

envi

rinae

F

Teq

uatr

ovi

rus

NC

_019399.1

OE

55

05

**

E

sch

eric

hia

ph

age

vB

_E

coM

_O

E5505

1

68

75

6

35.2

273

M

yovi

ridae

Tev

envi

rinae

F

Teq

uatr

ovi

rus

MK

373785.1

*N

o s

ubfa

mily i

s def

ined

acc

ord

ing t

o the

Inte

rnat

ional

Com

mitte

e on T

axonom

y V

iruse

s (I

CT

V).

Phag

e gen

us

is u

sed inst

ead. N

/A =

non

-applica

ble

. -

= s

ingle

ton,

no c

lust

er.

**N

ot fo

und in

IC

TV

dat

abas

e. C

lass

ific

atio

n a

ccord

ing t

o N

atio

nal

Cen

ter

for

Bio

tech

nolo

gy info

rmat

ion (

NC

BI)

.


97

Supplementary Table S2 | Overview of topological and branch length agreement

ape package phangorn package

Analysis Tree

comparison

Topological

distance

score

Robinson-

Foulds

distance

Symmetric

difference

Branch

score

difference

Path

difference

Sipho-

viridae

kmer10 vs.

kmer21 12.16553 324 323 - 2422.025

kmer10 vs. roary 15.71623 329 241 - 2683.053

kmer21 vs. roary 15.26434 333 348 - 2713.845

Myo-viridae

kmer10 vs.

kmer21 16.67333 346 346 - 2224.267

kmer10 vs. roary 12.56981 340 359 - 3640.188

kmer21 vs. roary 15.81139 346 365 - 3666.915

Single

genes

TLS vs. PP 3.38274 102 102 7.671297 272.736136

TLS vs. Exo 3.56119 112 124 7.215488 368.323228

PP vs. Exo 2.94990 120 132 6.026867 419.485399

kmer10 = tree based on kmer (10) presence/absence based on de novo assembled contigs. Kmer21 = tree based

on kmer (21) presence/absence based on de novo assembled contigs. Roary = tree based on gene presence/absence

within the full genomes. TLS = terminase large subunit. PP = portal protein. Exo = exonuclease.

The “Topological distance score” is calculated using the dist.topo function (score method).

The “Robinson-Foulds distance” is calculated using the RF.dist function.

The “symmetric difference” is calculated using the treedist function. It is similar to the Robinson-Foulds distance

and the Penny and Hendy’s distance.

The “Path difference” is calculated using the treedist function and is the difference in path lengths, counted as the

numbers of branches, between the pairs of tips


98

Su

pp

lem

en

tary

Ta

ble

S3

| O

verv

iew

of

ass

em

bly

sett

ings

an

d s

tati

stic

s

R

aw

da

ta

T

rim

min

g

Ass

em

bly

Ph

age

# o

f

rea

ds

Av

g.

len

gth

# o

f re

ad

s

aft

er t

rim

Av

g. le

ng

th

aft

er t

rim

# o

f m

atc

hed

red

s

Min

. co

nti

g

len

gth

Max. co

nti

g

len

gth

# o

f

con

tigs

N50

Phag

e 8

20854

212.9

20792

199.1

20792

51031

51031

1

51031

Phag

e 10

175980

212.0

175598

212.2

174572

380

167166

4

167166

Phag

e 1

1

23

71

28

1

97

.3

23

64

93

1

97

.6

23

58

00

5

03

169478

3

169478

Phag

e 1

5

78

81

6

18

4.6

7

83

77

1

85

.2

77

99

2

16

9392

169392

1

169392

Phag

e 17

28544

204.8

28.4

73

204.9

28300

250

44592

5

44592

Phag

e 18

122820

179.6

122040

180.3

121527

477

169391

2

169391

Phag

e 28

21654

230.5

21628

230.4

21376

254

52716

2

52716

Phag

e 3

0

16

20

94

1

87

.9

16

14

82

1

88

.3

16

06

23

2

53

169109

10

169109

Phag

e 4

7

32

58

2

19

8.2

3

24

62

1

98

.4

32

30

5

95

8

50105

2

50105

Phag

e 48

18862

211.8

18844

210.8

18665

51031

51031

1

51031

Phag

e 52

44154

208.7

44024

208.7

43542

373

50784

5

50784

Phag

e 5

3

37

15

4

21

0.5

3

70

72

2

10

.6

36

88

3

25

3

50582

2

50582

Phag

e 5

4

62

70

2

21

0.2

6

25

41

2

10

.3

62

21

3

25

1

51031

5

51031

Phag

e 55

233558

198.8

232979

199.0

231603

254

167131

8

167131

Phag

e 56

55786

209.9

55672

209.9

55010

50829

52716

2

52716

Phag

e 58

51386

209.7

51321

209.6

51069

256

44592

3

44592

Phag

e 5

9

16

27

0

23

5.4

1

62

58

2

35

.3

16

05

3

67

1

51031

2

51031

Phag

e 6

0

34

73

0

20

2.1

3

46

24

2

02

.5

34

43

5

86

237

86237

1

86237

Phag

e 61

219040

202.5

218539

202.5

217486

254

108138

6

108138

Phag

e 62

27242

201.9

27136

202.4

26952

87871

87871

1

87871

Phag

e 6

3

50

71

8

20

6.5

5

06

16

2

06

.5

50

41

6

49

132

49132

1

49132

Phag

e 6

4

44

87

2

20

9.0

4

47

58

2

09

.1

44

52

7

32

1

51031

2

51031

Phag

e 65

41890

210.1

41804

210.2

41612

51031

51031

1

51031

Phag

e 66

46548

208.5

46400

209.0

46135

257

89604

3

89604

Phag

e 68

65582

211.2

65499

211.0

65188

262

51029

2

51029


99

Su

pp

lem

en

tary

Ta

ble

S3

| C

on

tin

ued

R

aw

da

ta

Tri

mm

ing

A

ssem

bly

Ph

age

# o

f

read

s

Avg.

len

gth

# o

f re

ad

s

aft

er t

rim

Avg. le

ngth

aft

er t

rim

# o

f m

atc

hed

red

s

Min

. co

nti

g

len

gth

Max. co

nti

g

len

gth

# o

f

con

tigs

N50

Phag

e 6

9

48

63

2

21

4.9

4

85

32

21

4.8

4

79

24

2

78

50777

22

50777

Phag

e 7

0

45

03

2

21

2.0

4

49

34

21

2.1

4

47

11

2

50

44290

2

44290

Phag

e 71

24156

230.6

2

41

28

230.7

23787

415

51031

2

51031

Phag

e 72

32508

222.4

3

24

42

222.5

32100

253

51031

2

51031

Phag

e 7

3

42

83

2

22

6.7

4

27

52

22

6.8

4

24

07

4

6738

46738

1

46738

Phag

e 7

4

61

36

2

21

1.3

6

12

08

21

1.5

6

07

37

2

58

45171

4

45171

Phag

e 75

44444

208.3

4

43

54

208.3

44086

50445

50445

1

50445

Phag

e 76

43978

219.5

4

39

26

219.3

43522

504

50843

3

50843

Phag

e 77

59306

215.2

5

91

83

215.3

58920

51073

51073

1

51073

Phag

e 7

8

49

21

8

20

6.4

4

90

98

20

6.6

4

86

77

2

56

89644

2

89644

Phag

e 7

9

50

82

2

24

6.4

5

08

16

24

6.4

5

03

60

2

53

87100

7

87100

Phag

e 80

46730

207.1

4

66

08

207.2

46307

52703

52703

1

52703

D

efa

ult s

ett

ings

wa

s u

sed f

or C

LC

Genom

ics

Wo

rkbench d

e n

ov

o a

ssem

bly

, ex

cept f

or m

inim

um

co

ntig le

ngth

th

at w

as

chan

ged

to

250 b

p.


100

Su

pp

lem

en

tary F

igu

re S

1 |

Ro

ary

matr

ix b

ase

d o

n g

en

e li

st o

f S

iph

ovi

rida

e colip

hages.

Ph

ages

iso

late

d in

th

is s

tud

y a

re h

igh

ligh

ted. E

ach

co

lour re

pre

sents

a c

lust

er:

Clu

ster A

(b

lue),

clu

ster B

(gre

en

), a

nd

clu

ster

C (re

d).

Clu

ster A

su

bclu

ster

s in

clu

de

A1

(ligh

t b

lue),

A2

(b

lue),

an

d A

3 (d

ark

blu

e).


101

Su

pp

lem

en

tary F

igu

re S

2 |

UP

GM

A tre

e b

ase

d o

n r

oary

matr

ix o

f S

iphovir

idae c

olip

hages.

Ph

ages

iso

late

d in

th

is s

tud

y a

re h

igh

ligh

ted. E

ach c

olo

ur re

pre

sents

a

clu

ster:

Clu

ster A

(b

lue),

clu

ster B

(gre

en

), a

nd

clu

ster C

(re

d).

Clu

ster A

su

bclu

ster

s in

clu

de

A1

(ligh

t b

lue),

A2

(b

lue),

an

d A

3 (d

ark

blu

e).


102

Su

pp

lem

en

tary F

igu

re

S2

| R

oary

matr

ix b

ase

d o

n g

en

e list

of

Myovir

idae c

olip

hages.

Ph

ages

iso

late

d in

th

is s

tudy a

re h

igh

ligh

ted. E

ach

colo

ur

rep

rese

nts

a c

lust

er:

Clu

ster D

(o

ran

ge),

clu

ster E

(p

urp

le),

and

clu

ster F

(bro

wn

).


103

Su

pp

lem

en

tary F

igu

re

S4

| U

PG

MA

tre

e b

ase

d o

n ro

ary

matr

ix o

f M

yovir

idae c

olip

hages.

Ph

ages

iso

late

d in

th

is s

tud

y a

re h

igh

ligh

ted

. Each

co

lou

r re

pre

sents

a

clu

ster:

Clu

ster D

(o

ran

ge)

, clu

ster E

(p

urp

le),

and

clu

ster F

(b

row

n).


104

Supplementary Figure S5 | Maximum likelihood tree based on the nucleotide sequence of the phage portal

protein. Phages were grouped together into six clusters: A-F, according to phage family and subfamily. Cluster A

and B: Siphoviridae, Tunavirinae, cluster C: Siphoviridae and Tequintavirus genus, cluster D: Myoviridae,

Ounavirinae, and cluster E and F: Myoviridae, Tevenvirinae. Cluster A was divided into three subclusters: A1,

A2 and A3. Subcluster A2 was divided in two. The tree was constructed using the MEGA X software. The

percentage of threes in which the associated taxa clustered together is shown next to the branches. The tree is

drawn to scale, with branch lengths measured in the number of substitutions per site. The analysis involved 62

nucleotide sequences.

A1

A2

A2

A3


105

Supplementary Figure S6 | Maximum likelihood tree based on the nucleotide sequences of phage exonucleases.

The tree comprised six clusters: A-F, according to phage family and subfamily. Cluster A and B: Siphoviridae,

Tunavirinae, cluster C: Siphoviridae and Tequintavirus genus, cluster D: Myoviridae, Ounavirinae, and cluster E

and F: Myoviridae, Tevenvirinae. Cluster A was divided into three subclusters: A1, A2 and A3. Cluster C and F

were found in two and three copies, respectively. The tree was constructed using the MEGA X software. The

percent of data coverage for internal nodes is indicated. The tree is drawn to scale, with branch lengths measured

in the number of nucleotide sequence substitutions per site. The 20 reference phages included for comparison.

The analysis included 73 nucleotide sequences.

A1

A2

A3


106

Comparative genomics (Supplementary Figure S7-14)

General description

Phage genome sequences were compared for each cluster (cluster A-F) using the

progressiveMauve software. Boxes with same colours represent LCBs, indicating homologous

DNA regions shared by two or more genomes without sequence rearrangements. LCBs

indicated below the horizontal black line represent reverse compliments of the reference LCB

(reference genome is marked with a blue square). The height of the similarity profile within

the LCBs corresponds to the average level of conservation in that region of the genome

sequence. White boxes below the horizontal black line represents annotated genes in the

reference sequences included. The terminase large subunit encoding genes is indicated with a

black square in each genome sequence.


107

Supplementary Figure S7 | Comparative genomics of subcluster A1 phages. Genome sequences of 17

Siphoviridae subcluster A1 phages were compared. 16 LCBs were identified.


108

Supplementary Figure S8 | Comparative genomics of subcluster A2 phages. Genome sequences of seven

Siphoviridae subcluster A2 phages were compared. Seven LCBs were identified.

Supplementary Figure S9 | Comparative genomics of subcluster A3 phages. Genome sequences of four

Siphoviridae subcluster A3 phages were compared. Four LBCs were identified.


109

Supplementary Figure S10 | Comparative genomics of cluster B phages. Genome sequences of six Siphoviridae

cluster B phages were compared. Six LCBs were identified.


110

Supplementary Figure S11 | Comparative genomics of cluster C phages. 10 Genome sequences of 10

Siphoviridae phages were compared. 6-10 LCBs were identified in each genome. Repeat-rich regions are indicated

with salmon-coloured bars next to the white annotation boxes.


111

Supplementary Figure S12 | Comparative genomics of cluster D phages. Genome sequences of 19 Myoviridae

cluster D phages were compared. 14-17 LCBs were identified for each phage genome.


112

Supplementary Figure S13 | Comparative genomics of cluster E phages. Genome sequences of 18 Myoviridae

cluster E phages were compared. 16-17 LCBs were identified for each phage genome.


113

Supplementary Figure S14 | Comparative genomics of F subcluster phages. Genome sequences of 13 Myoviridae

cluster F phages were compared. Five LCBs were identified for each phage genome.


115

3.2

Classification of in vitro phage-host population growth dynamics

Patricia E. Sørensen1,2, Duncan Y. K. Ng3, Luc Duchateau4 , Hanne Ingmer5, An

Garmyn1, and Patrick Butaye1,2

1 Department of Pathobiology, Pharmacology and Zoological Medicine, Ghent University, Merelbeke,

Belgium


Indies

3 Department of Bacteria, Parasites and Fungi, Statens Serum Institut, Denmark

4 Biometrics Research Center, Ghent University, Belgium


Published in Microorganisms 2021, 9 (2470)



116

Abstract

The therapeutic use of bacteriophages (phage therapy) represents a promising alternative to

antibiotics to control bacterial pathogens. However, the understanding of the phage-bacterium

interactions and population dynamics seems essential for successful phage therapy

implementation. Here, we investigated the effect of three factors, phage species (18 lytic E.

coli-infecting phages), bacterial strain (10 APEC strains), and multiplicity of infection (MOI)

(MOI 10, 1, and 0.1) on the bacterial growth dynamics. All factors had a significant effect, but

the phage appeared to be the most important. The results showed seven distinct growth patterns.

The first pattern corresponded to the normal bacterial growth pattern in the absence of a phage.

The second pattern was complete bacterial killing. The remaining patterns were in -between,

characterised by delayed growth and/or variable killing of the bacterial cells. In conclusion,

this study demonstrates that the phage-host dynamics is an important factor in the capacity of

a phage to eliminate bacteria. The classified patterns show that this is an essential factor to

consider when developing a phage therapy. This methodology can be used to rapidly screen

for novel phage candidates for phage therapy. Accordingly, the most promising candidates

were phages found in Group 2, characterised by growth dynamics with high bacterial killing.


117

Introduction

Bacteriophages (phages) are viruses that specifically infect bacteria. They are estimated to be

the most abundant organisms on Earth (~1031 entities) and play a major role in shaping the

microbial communities [1, 2]. Phages are unable to replicate independently of a susceptible

bacterial host, and their host range is determined by a combination of various factors, including

host-binding protein specificity and bacterial phage-resistance mechanisms [3, 4]. Bacteria can

readily evolve resistance to phage infections though different mechanisms, which can result in

distinct resistance phenotypes [5]. These can differ in whether the resistance is partial or

complete, in the level of fitness cost associated with resistance, and in whether the mutation

can be countered by the infecting phage. Consequently, these differences determine the effect

of the phage infection on the bacterial population dynamics and the resulting community

structure [6, 7].

The therapeutic use of phages (phage therapy) represents an urgently needed alternative or

supportive antibacterial agent to antibiotics to control bacterial pathogens [8–11]. However,

the outcome of phage therapy is still difficult to predict [12]. Several factors can be involved,

both animal host and bacterial host related as well as phage related. Whether phages are able

to infect and kill a susceptible target bacterial population at a specific site depends on the

change in phage densities in different tissues of the host (pharmacokinetics (PK)) and the

population dynamics of the phage-bacterial interaction (pharmacodynamics (PD)) [7].

Understanding the phage-bacterium interactions and population dynamics has been shown to

be essential for more reliable in vitro and/or in vivo outcomes, and thereby, essential for

successful phage therapy development and application [7, 13, 14]. Several models have been

developed to predict the behaviour and dynamics of phage-bacteria populations [15–18]. These

models include parameters such as bacterial growth and mutation rate, phage adsorption rate,

burst size, latent period, and virulence, as well as multiplicity of infection (MOI) [7, 18]. While

no single model to date has been able to capture all aspects of the complex phage -host

interactions, together, suitable models can be selected to predict and explain basic behaviours

of the population dynamics, as well as identify the dominant factors that contribute the

dynamics [7, 15, 17, 18]. The dynamics of in vitro phage-host interactions may differ from

those observed in vivo. This is similar to what is seen also with antibiotics [19, 20]. Still, in

vitro observations constitute a necessary initial step in understanding and predicting phage -

host population dynamics in in vivo settings [7]. Several models focus on dynamics of only an

individual phage and single host [17, 18], and can fail to report what defines the dynamic(s).


118

This study investigates the in vitro growth dynamics of a group of Escherichia coli (E. coli)-

infecting phages (coliphages) and multi-drug resistant avian pathogenic E. coli (APEC). We

define the specific patterns observed using a multiple parameter-based approach and estimate

the parameter(s) (phage type, APEC strains, and MOI) that are key to each pattern. APEC (with

O-serogroups O1, O2, or O78) was chosen as the bacterial model due to the significant problem

it represents to poultry worldwide [21, 22]. This pathogen causes a large range of extraintestinal

infections, collectively referred to as colibacillosis, which are becoming harder to treat with the

increasing resistance of APEC to different classes of antibiotics [23, 24].

Methods

Bacterial strains and growth conditions

The 10 APEC strains are part of an in-house collection that were isolated from clinical poultry

faeces samples suspected of APEC infection. Samples were collected in Belgium during 2013-

2014 by the Animal Health Care Flanders (Torhout, Belgium). Strains were grown in LB broth

(Miller) (Sigma-Aldrich, Saint Louis, MO, USA) or on LB agar supplemented with 1.5%

bacteriological agar no. 1 (w/v) (Oxoid, Basingstoke, UK) overnight (16-18 h) at 37 °C unless

stated otherwise. Broth cultures were incubated with shaking (120 rpm). Strains were stored at

-80 °C in LB broth supplemented with 15% glycerol (Sigma-Aldrich, Saint Louis, MO, USA).


Bacterial genomic DNA was extracted using Qiagen’s DNeasy Blood and Tissue Kit (Qiagen,

Hilden, Germany), with the subsequent library construction using the Nextera XT Kit

(Illumina, Little Chesterford, UK), and sequenced using a 300-cycle kit on the Illumina

NextSeq platform according to the manufacturer’s instructions.

Bacterial genome analysis

The bifrost platform (https://github.com/ssi-dk/bifrost) (accessed on 15 January 2021), v1.1.0,

was used for quality control validation of the raw reads data. The raw reads were de novo

assembled using SPAdes v3.12.1 [25] and MLST typed with the MLST command-line tool

(https://github.com/tseemann/mlst) (accessed on 15 January 2021), v2.19. The serotypes were

predicted using SerotypeFinder, v2.0 [26]. ABRicate v1.0.1

https://github.com/ssi-dk/bifrost

https://github.com/tseemann/mlst


119

(https://github.com/tseemann/abricate) (accessed on 18 June 2021) with default options was

used to screen the assembled contigs for antimicrobial resistance genes with ResFinder [27]

and the Comprehensive Antibiotic Resistance Database (CARD) [28]. Virulence genes were

identified using ABRicate with Ecoli_VF database data. Prophage regions were identified

using the PHAge Search Tool Enhanced Release (PHASTER) tool [29]. The Rapid Annotation

using Subsystem Technology (RAST) server and the SEED viewer, v2.0 [30], were used to

identify coding sequences (CDSs) and for initial annotation of the APEC draft genomes.

Bacteriophage isolation, purification, and enumeration

A total of 18 lytic coliphages were used in this study. These were selected from the collection

based on their genomic diversity. Phages were isolated from poultry faecal material using E.

coli laboratory strain K514, sequenced and identified as described before [31]. Phage stocks

were stored at titers of 1.2x107 to 4.5x1010 plaque forming units (PFU)/ml at 4 °C. Working

stocks used for phage infectivity and phage-host growth experiments were kept at titers of

~1010 PFU/ml.

Phage infectivity and phage-host growth dynamics

Bacterial overnight cultures were used, and the cell concentration was adjusted to ~108 CFU/ml

for each experiment. Bacterial solutions were inoculated with phage, yielding initial MOIs of

10, 1, or 0.1. All bacterial reduction curves were generated using 96-well plates with working

volumes of 200 µl. Experiments were performed in triplicate. For the experiments of the

susceptible combinations, the experiment was performed on a duplicate plate at another time.

A well of phage-free bacterial culture and a well of bacteria-free phage culture were included

on every plate as control experiments in addition to one media blank for reference. The optical

density (OD) was measured for a wavelength of 600 nm (OD600) with the Thermo Fisher

Scientific Multiskan GO Microplate Spectrophotometer, v1.01.12, and the data were recorded

using SkanIt software, v6.0.2.3. The OD600 was measured with fast measurement mode and

no pathlength correction or use of transmittance, and the measurements were taken

immediately after inoculation and then at regular intervals of 30 minutes afterward for 22 hours.

The incubation temperature was 37 °C and shaking was continuous at a medium speed.

Growth curves were obtained by plotting OD600 values after baseline adjustment against time.

Phage infectivity was defined based on endpoint measurements. Successful phage infection


120

was defined as OD600: <0.2, somewhat successful infection was defined as OD600: 0.2 -0.5,

and failed phage infection as OD600: >0.5. The phage-host growth dynamics were assessed

based on measurements throughout the experiment.

Assessing the effect of phage species, APEC strain and MOI

The two different response variables, PhageScore [32] and local virulence score [16] were

derived to assess the effect of phage species, APEC strain, and MOI on the growth dynamics.

For the PhageScore method, the area under the growth curve (AUC) was determined for the

complete study period for each treatment combination and for the corresponding control, i.e.,

without phage, and the ratio of the difference between the control and the treatment over the

control (multiplied with 100) was calculated. For the local virulence score, the same was done,

but only measurements until the stationary phase in the control group were taken into

consideration, i.e., until the timepoint before the maximum OD value.

A fixed effects model with a normally distributed error term was used, and phage, bacterium,

and MOI, as well as all two-way interactions, were included in the model. F-tests were

performed at the 5% significance level to assess the effects of the different factors. Finally, the

Pearson correlation coefficient between the PhageScore and the virulence score was calculated.

Classification of phage-bacterium growth dynamics

To classify the growth dynamics of phage-host interactions based on the OD measurements,

we applied two statistical data mining techniques: non-metric multidimensional scaling

(NMDS) ordination and principal component analysis (PCA), with hierarchical agglomerative

clustering. A NMDS ordination plot (Euclidean distance) using the vegan package in R

(http://www.R-project.org/) (accessed on 18 April 2021) [33] was applied to quantify and

visualise the pairwise dissimilarity between samples of each timepoint. The stress value was

determined to access how well the data were transformed. A stress value between 0.02 and

0.01 was considered an acceptable fit, and <0.01 was considered a good fit. Metadata was

included using the envfit function to determine the effect of each factor (phage species,

bacterial strain, and MOI). Using the factoextra package the right number of groups (of the

growth dynamics pattern) was determined using bootstrap values = 100 and hierarchical

agglomerative clustering (“ward.D” method). The results were visualised using the ggplot2

package. A PCA was performed as verification (sanity check) of how much of the variability

http://www.r-project.org/


121

in the data is explained by each factor and how much of the total variability is captured.

Subsequent subclustering of the designated groups was performed as described above for the

complete dataset.

Repeatability of group assignments

The dplyr package in R was used to determine if replicates of the same phage-bacterium-MOI

combination were placed in the same group and subgroup.

Results

APEC strains

Whole-genome sequencing (WGS) of the bacterial genomes yielded a total of 2,673,858-

4,825,608 paired-end reads for each of the 10 isolates, with an average coverage of 77-142-

fold. The characteristics based on the WGS analysis are summarised in Table 1. The strains

belonged to one of four serotypes and MLST sequence type (ST): Strains B1, B4, and B8 had

serotype O1:H7 (ST95); strains B5 and B10 had serotype O2:H5 (ST355); strains B2 and B3

had serotype O78:H4 (ST117); and strains B6, B7, and B9 had O78:H9 (ST23). Each bacterial

genome comprised between 5 and 13 prophage regions, including a total of 40 different

prophages (Supplementary Table S1).

Table 1 | Avian pathogenic Escherichia coli (APEC) strain characteristics

Strain

name

APEC

isolate

Genome

size (kb) Serotype

MLST

ST

G+C

content

(%)

CARD

genes

Virulence

associated

genes

Prophage

regions

B1 AM621 5274.2 O1:H7 95 50.6 52 225 11

B4 AM635 5015.3 O1:H7 95 50.6 44 224 6

B8 AM646 5025.1 O1:H7 95 50.6 44 223 6

B5 AM639 5044.7 O2:H5 355 50.6 46 200 6

B10 AM650 4984.4 O2:H5 355 50.6 43 210 5

B2 AM631 5144.8 O78:H4 117 50.6 46 199 11

B3 AM632 5160.1 O78:H4 117 50.6 50 180 13

B6 AM642 4863.7 O78:H9 23 50.6 51 178 9

B7 AM644 5056.7 O78:H9 23 50.6 51 187 12

B9 AM648 5037.0 O78:H9 23 50.5 50 188 11

ST = Sequence type. CARD: Comprehensive Antibiotic Resistance Database


122

Coliphages

The coliphages used in this study were previously characterised [30]. The characteristics are

summarised in Table 2. The phages belonged to one of seven different genera. Tequatrovirus

phages included P1, P2, and P9. Mosigvirus phages included P3, P5, and P6. Guelphvirus

phages included P4, P13, and P14. Hanrivervirus phages included P7, P8, and P16.

Felixounavirus phages included P10, P12, P17, and P18. Tequintavirus phages included P11.

Warwickvirus phages included P15.

Table 2 | Coliphage characteristics

Na

me

Phage

name

Genome

size (kb) Phage family

Phage

subfamily Phage genus Accession no.

P10 Phage 60 86.2 Myoviridae Ounavirinae Felixounavirus SRX8360069




P4 Phage 17 45.9 Drexlerviridae Braunvirinae Guelphvirus SRX8360091



P7 Phage 53 50.8 Drexlerviridae Tempevirinae Hanrivervirus SRX8360063



P3 Phage 15 169.6 Myoviridae Tevenvirinae Mosigvirus SRX8360082



P1 Phage 10 169.0 Myoviridae Tevenvirinae Tequatrovirus SRX8360061



P11 Phage 61

109.9 Demerecvirida

e

Markadamsvir

inae Tequintavirus SRX8360070

P15 Phage 76 51.9 Drexlerviridae Tempevirinae Warwickvirus SRX8360086

Phage classification according to current (16 September 2021) International Committee on Taxonomy of

Viruses (ICTV) taxonomy.

Phage infectivity

The infectivity of the 18 coliphages against each of the 10 APEC strains at MOI 10, 1 , and 0.1

is shown in Figure 1. The levels of infectivity of the tested coliphages were highly variable,

varying between 0% and 100% of the tested APEC strains. Variations in the degree of infection

(successful, somewhat successful, or failed) were detected, with the most successful infections

at MOI 10 and least at MOI 0.1. Phage P1 was shown to be the most infective , as it was the


123

only one able to infect all APEC strains at all MOI tested, except for B7-MOI 0.1. Phages P2,

P3, and P9 were able to infect nine of the APEC strains, excluding B4, B4, and B1, respectively.

Phage P5 was able to infect eight of the APEC strains, excluding B1 and B4. Phage P8 was

able to infect six of the APEC strains, not including B1, B4, B6, and B7. Phages P4, P6, and

P7 were able to infect five of the APEC strains, excluding B1, B4, B6, B7, and B8. The nine

phages P10-18 were not able to infect any of the 10 APEC strains.

Figure 1 | Infectivity of the tailed coliphages based on the endpoint OD. P = phage. B = bacterium. MOI =

multiplicity of infection. Ratio = phage:APEC. OD was measured at 600 nm. (OD600). Dark blue = final OD600:

<0.2, light blue = final OD600: 0.2-0.5, and red = final OD600: >0.5. The final OD was determined based on the

average of three-nine replicates.

Assessing the effect of phage species, APEC strain, and MOI

A total of 2869 phage-host combination experiments were performed. The average local

virulence score and its standard error for each phage-bacterium combination are shown in

Figure 2. All factors were significant, i.e., MOI (F2,2495 = 938.8, p < .0001), Phage

(F17,2495=1095.8, p < .0001) and Bacterium (F9,2495=459.5, p < .0001), and the interactions too

but to a lesser extent, i.e., Phage-Bacterium (F153,2495=42.8, p < .0001), Phage-MOI

(F34,2495=10.7, p < .0001) and Bacterium-MOI (F18,2495=9.7, p < .0001). The phage effect was

thus more pronounced than the bacterium effect. Based on the virulence scores, the greatest


124

reduction in bacterial growth of the six strains B1, B2, B4, B6, B7, and B8 was observed with

phage P1 and with phage P5 for the four strains B3, B5, B9, and B10.

Figure 2 | Virulence score average by bacteria and phage. A higher virulence score correlates with a higher

virulence of the phage and higher bacterial growth reductions. The standard error is shown. A total of 2869 phage-

host combination experiments were included.

The average PhageScore and its standard error for each phage-bacterium combination are

shown in Figure 3. All factors were significant, i.e., MOI (F2,2495 = 447.9, p < .0001), Phage

(F17,2495 = 693.6, p < .0001) and Bacteria (F9,2495 = 364.6, p < .0001), and the interactions too

but to a lesser extent, i.e., Phage-Bacteria (F153,2495 = 38.1, p < .0001), Phage-MOI (F34,2495 =

13.0, p < .0001) and Bacteria-MOI (F18,2495 = 12.5, p < .0001). The phage effect was therefore

also more pronounced than the bacterium effect when looking at the PhageScore. Based on the

PhageScore, the greatest reduction in bacterial growth of the four strains B1, B4, B6, and B8

was observed with phage P1, and with phage P5 for the four strains B5, B7, B9, and B10. The

greatest reduction of B2 and B3 growth was observed for phages P9 and P6, respectively.


125

Figure 3 | PhageScore average by bacteria and phage. A higher PhageScore correlates with a higher efficiency of

the phage and higher bacterial growth reduction. A total of 2869 phage-host combination experiments were

included. The standard error is shown.

The Pearson’s correlation coefficient between the local virulence score and the PhageScore

was equal to 0.94.

Classification of phage-bacterium growth dynamics

Phage-host growth dynamics were classified based on the OD measurements from a total of

2,729 phage-host combination experiments. The NMDS analysis grouped the data into three

different groups (Figure 4). Group 1 comprised resistant phage-host combinations

characterised by bacterial growth, Group 2 comprised fully susceptible combinations

characterised by bacterial killing, and Group 3 comprised in-between combinations. A two-

dimensional plot was considered appropriate, as the generated stress value was below 0.01

(0.007). Growth dynamics curves associated with each combination experiment and phage-free


126

control culture are shown in Supplementary Figure S1. PCA captured a total of 97% of the

variance and showed similar groupings (Supplementary Figure S2).

Figure 4 | Grouping based on non-metric multidimensional scaling (NMDS) analysis. The analysis resulted in

three groups. Group 1 comprises resistant combinations with bacterial growth, as well as phage-free controls.

Group 2 comprises susceptible combinations with bacterial killing. Group 3 comprises in -between combination.

A total of 2,869 entities are shown on the plot. Ellipses indicate a 95% confidence level based on a multivariate

t-distribution. Stress < 0.01 = a good fit.

NMDS on a two-dimensional graph was applied to investigate how the factors (bacterial strain,

phage species, and MOI) affect the groupings (Figure 5). The 10 phages P4 and P10-18 were

associated with bacterial growth (Group 1), of which P11 had the greatest association (similar

to the phage-free controls), followed by P17, P4, P16, P13, P18, P10, P15, P12, and P14 (Figure

5A). Accordingly, the greatest bacterial growth was associated with the Demerecviridae phage

belonging to the Tequintavirus genus. Myoviridae phages belonging to the Felixounavirus

genus and Drexlerviridae phages belonging to the Guelphvirus or Warwickvirus genera were

also associated with bacterial growth. Phages associated with more bacterial killing (Groups 2

and 3) included the eight phages P1-3 and P5-9, of which P1 had the strongest association,

followed by P5, P9, P2, P3, P8, P6 and P7. A few P4-combinations were also found in Group

3.


127

Figure 5 | A non-metric multidimensional scaling (NMDS) plot of the factors driving the grouping. A) Effect of

the phage type, including phage-free controls (empty). B) Effect of the bacterial strain. C) Effect of the multiplicity

of infection (MOI). MOI 0 represents phage-free controls. A total of 2,869 entities were divided into three groups.

Group 1 represents resistant combinations/bacterial growth. Group 2 represents susceptible

combinations/bacterial killing. Group 3 represents in-between combinations. Arrows indicate the strength and

direction of the correlation. Ellipses indicate a 95% confidence level based on a multivariate t-distribution. Stress

< 0.01 = a good fit.


128

Accordingly, the greatest bacterial killing was associated with Myoviridae phages belonging

to the Tequatrovirus genus. Myoviridae phages belonging to the Mosigvirus genus were also

associated with bacterial killing, as well as two out of three Hanrivervirus phages

(Drexlerviridae family). The five bacterial strains B1, B4, B6, B7, and B8 were associated with

bacterial growth (Group 1). This included all three O1-serotype strains (B1, B4 and B8), as

well as two O78-serotype strains (Figure 5B). The other five strains: B2, B3, B5, B9, and B10,

were associated with more bacterial killing and included two O2-serotype strains (B5 and B10)

and three O78-serotype strains. The overall effect of the bacterial factor on the plot was less

compared to the effect of the phage factor. A high MOI (MOI 10) was associated with more

bacterial killing, and a low MOI (MOI 0.1) was associated with less killing (Figure 5C).

Based on the NMDS analysis, Group 3, including 419 entries, was divided into five subgroups

(Figure 6). Only combinations with phages P1-P9 were found. The growth curves associated

with each subgroup are shown in Supplementary Figure S3.

Figure 6 | Subgrouping of Group 3 based on non-metric multidimensional scaling (NMDS). Group 3 (n = 419)

was divided into five subgroups. Subgroups 3.1 (n = 167), 3.2 (n = 95), and 3.3 (n = 43) represent combinations

with initial bacterial killing followed by exponential bacterial growth and stationary phase. Subgroups 3.4 (n =

66) and 3.5 (n = 48) represent combinations with initial bacterial growth followed by stationary phase. Ellipses

indicate a 95% confidence level based on a multivariate t-distribution. Stress < 0.02 = an acceptable fit.


129

NMDS on a two-dimensional graph was applied to investigate how the factors (bacterial strain,

phage species, and MOI) affect the Group 3 groupings (Figure 7).

Figure 7 | Non-metric multidimensional scaling (NMDS) plot of the factors driving the Group 3 subgrouping. A)

Effect of the phage type. B) Effect of the bacterial strain. C) Effect of the multiplicity of infection (MOI). Arrows

indicate the strength and direction of the correlation. Ellipses indicate a 95% confidence level based on a

multivariate t-distribution. Stress < 0.02 = an acceptable fit.


130

The association of phage P3 was in the direction of subgroup 3.1 ; however, the association was

weak. Phages P1 and P5 had a strong association with subgroup 3.2, while P2 and P3 had a

weak association. P5 association was in the direction towards subgroup 3.3. The remaining five

phages, including P4, P6, P7, P8, and P8, were associated with subgroup 3.4. P6 was the only

phage with a strong association (Figure 7A). Bacterial strains B5 and B10 (O2 serotype) had a

strong association with subgroup 3.1, and B2 (O78) had a weak association. B1 (O1), as well

as O78 strains B6, B7, and B9 had an association with subgroup 3.2. B6 was the only strain

with a strong association, as well as the only strain driving the plot towards subgroup 3.3. No

strains were driving the plot towards subgroup 3.4. B3 was found in this subgroup but showed

a weak/no association. O1 strains B4 and B8 had a strong association with subgroup 3.5 (Figure

7B). The MOIs did not have a strong association with any of the subgroups (Figure 7C).

Description of defined growth dynamics patterns

Based on the bacterial growth (OD) detected in the 2,869 phage-host combination experiments,

three different growth dynamics patterns groups, including five subgroups, were defined

(Figure 8). The groups included: 1) Combinations with a fully resistant bacterial growth

pattern; 2) Phage-host combinations with a fully susceptible pattern, showing minimal or no

bacterial growth; and 3) Combinations with one of five in-between patterns characterised by

delayed growth, lower killing, or variable killing of the bacterial cells. For all dynamics patterns

except Group 1, the phage effect on the bacterial growth kinetics was observed within only a

few hours of incubation. When the stationary phase was reached, the bacterial density remained

stable throughout all the co-culturing experiments.

Group 1 (bacterial growth): The bacterial growth continued to increase during the first 7 h of

incubation until the cultures reached the stationary phase. The final OD600 was ~0.7. The

pattern observed showed logistic growth as under standard conditions without phage present

and could be explained by the presence of naturally phage-resistant strains, where the phage is

unable to infect and has no effect on the growth. Group 2 (high level of bacterial killing): The

bacteria were lysed and never recovered. A single small peak of bacterial growth followed by

bacterial killing was observed in some cases but was not reflected in the average OD growth

curve. Group 3.1 (initial bacterial killing followed by bacterial growth): Prolongation of the

lag phase with no or low bacterial growth for ~9 hours was observed , followed by a slow

increase in bacterial growth until the stationary phase was reached. The final OD600 was ~0.45.


131

The greatest variations were seen in this subgroup (Supplementary Figure S3). Group 3.2

(initial bacterial killing followed by bacterial growth): Prolongation of the lag phase with no

or low bacterial growth was observed, followed by a sharp increase in the bacterial density

after ~7 hours of incubation before reaching the stationary phase. The final OD600 was ~0.5.

Figure 8 | Growth dynamic patterns of the coliphage-APEC co-culture combinations. Patterns are defined based

on the average OD value for each timepoint for each group.

Group 3.3 was characterised in a similar way as Group 3.2, except for a higher final OD600

of ~7.5, the highest observed for the seven different patterns. Group 3.4 (impaired bacterial

growth): Increase of the bacterial growth was observed during the first ~9.5 hours of incubation

until the cultures reached the stationary phase. Impaired growth was observed compared to the

Group 1 and Group 3.5 patterns with a final OD600 of ~0.26. Group 3.5 was similarly

characterised by impaired bacterial growth. Increase of the bacterial growth was observed

during the first ~12 hours of incubation until the cultures reached the stationary phase. Impaired

growth was observed compared to the Group 1 patterns, with a final OD600 of ~0.45 (similarly

to Group 3.1).

Repeatability of group assignments

All replicates, including the phage-bacteria-MOI combinations with phage P10-P18, were

found in same NMDS group (Group 1). Some discrepancies in the group assignments were


132

seen for replicates of specific combinations, including eight phage P1-combinations, 11 P2-

combinations, three P3-combinations, nine P4-combinations, 12 P5-combinations, seven P6-

combinations, three P7-combinations, seven P8-combinations, and 10 P9-combinations

(Figure 9). Most discrepancies included combinations with replicates grouped in Group 2

(bacterial killing) and Group 3.1 (in-between and bacterial killing), replicates grouped in Group

2 and Group 3.2/3.3 (initial bacterial killing followed by bacterial growth), or replicates

grouped in Group 1 (bacterial growth) and Group 3.4/3.5 (in between dynamics and bacterial

impaired growth) (Supplementary Tables S2 and S3).

Figure 9 | Overview of the discrepancies of the grouping of the phage-APEC-MOI combination replicates. P =

phage. B = bacterium. MOI = multiplicity of infection. White = no discrepancies. Black = one or more

discrepancies. Six-nine technical replicates were included for each combination. Combinations with P10-18 were

not included, as no discrepancies were detected.

Discussion

It has become clear that successful phage therapy development and application, among others,

depend on an understanding of the phage-host interactions and population dynamics [34]. This

study presents the classification of bacterial growth dynamics in the presence of lytic phages

using two statistical data mining techniques: NMDS and PCA. The use of OD measurements

represents a fast and data rich screening method for in vitro phage-host growth dynamics [35].

This approach captures the ongoing dynamics and produces quantitative high-throughput data

to determine the phage-host range, phage virulence or infectivity, and bacterial phage

resistance development [16, 32, 36]. These factors can be important as pharmacodynamic (PD)


133

parameters and include also a part of the pharmacokinetics (PK) as it assesses the potential

increase of the treatment dose. This would not be the case when relying on a single endpoint

measurement. However, it is understood that ODs do not differentiate between viable and dead

cells, and as such, there was no exact link between the OD values and bacteria viability.

Nevertheless, it is a good proxy for estimating bacterial numbers. The repeatability of the

NMDS grouping was shown to be acceptable. Fully natural phage-resistant combinations

(Group 1) were clearly identified. Most discrepancies observed between groupings of the

replicates from the same phage-APEC-MOI combinations can be explained by grouping cut-

off values. Replicates grouped in Group 1 and Group 3.4/3.5 included dynamics characterised

by higher or lower bacterial growths. Replicates grouped in Group 2 and Group 3.1, included

dynamics characterised by the initial bacterial killing, with no or low subsequent growth.

The optimal number of clusters/groups varied depending on the method used. In this study, we

chose five clusters. However, subgroups 3.2 and 3.3 and subgroups 3.4 and 3.5 were relatively

similar and could potentially be combined, resulting in only three Group 3 subclusters, and this

would also create fewer discrepancies in the group allocations. The discrepancies due to

biological variations can be expected due to the spontaneous emergence of phage-resistant

variants after varying incubation time (Group 2 vs. Groups 3.1, 3.2, and 3.3). It is also possible

for a very small (partially) resistant sub-population of bacteria to be naturally present in the

culture at the start of the experiment [17].

Various factors affecting the phage PK/PD have been described using mathematical and

experimental models [15, 17, 18, 37, 38]. In this study, the influence of the factors (phage type,

bacterial strain, and MOI) on the observed growth patterns was determined. Previous studies

have highlighted the MOI influence on phage therapy, and recently, a fast microtiter plate assay

for determination of the optimum MOI for a coliphage was further described [39]. However,

in this study, we found the MOI to have a less significant effect on the phage -host growth

dynamics outcome compared to the phage species. Furthermore, a recent study suggested that

the description of MOI alone is not sufficient, as the concentration, particular to the bacteria,

can significantly affect the results [40]. Therefore, in this study, the MOI at all the tested values

was based on a fixed bacterial concentration.

A quantitative assessment of the phage lytic activity using the virulence score and PhageScore

across a large dataset allowed direct comparisons of individual phages. In contrast to a single

OD endpoint measurement and the well-established plaque assay, these methods (virulence


134

score and PhageScore) captured the dynamics of phage infection, including bacterial

(re)growth after prolonged growth inhibition or lysis. Additionally, compared to the overlay-

based efficiency of plating assays and direct spot testing, these methods represent an accurate

and less cumbersome and time-consuming and do not depend on the subjectivity and/or

experience of the observer [41]. However, upscaling of this approach depends on the

availability of a high-throughput plate reader. In accordance with previous findings, we found

the two methods highly comparable, showing similar properties/values for the studied phages

[32, 42]. In this study, whenever the bacteria were able to grow and reach the stationary phase,

the growth remained stable throughout all the experiments. If a second peak would have

appeared or the growth would have started to decrease (after reaching the stationary phase), the

comparability of the local virulence score and the PhageScore would be reduced. In this study,

we only analysed the growth dynamics of co-cultures of a single phage type and bacterial strain.

However, both the virulence score and PhageScore have previously been used to compare

phage combinations for use in phage therapy cocktails [32, 41]. In future studies, the inclusion

of mixed phage cultures (phage cocktails), preferably targeting different host receptors, may

provide further indications of their potential as therapeutics against pathogenic target bacteria

[17]. Accordingly, for future applications, the inclusion of phage, as well as bacterial traits,

may be required for classification.

Given their great abundance and diversity, multiple candidate phages might be available to

infect a target host; yet, we still lack a better understanding of which phage would perform best

[43]. One approach to identify the cause(s) of treatment success is to compare the

characteristics of phages with high success rates with those of phages with low success rates.

The characteristics differing between these two groups of phages become candidates for

causation. In this study, Tequintavirus phages (associated with bacterial growth) and

Tequatrovirus phages (associated with bacterial killing) would be great candidates for

comparisons. Accordingly, the inclusion of phage characteristics, such as the phage receptor,

adsorption rate, latency period, burst size, and virion size, may provide further explanation of

the phage-host dynamics and may help predict the phage therapy efficacy [44–46].

Phage therapy is, by its nature, a strongly selective treatment [17]. Accordingly, when selecting

phages for therapeutical application, the emergence of phage-resistant bacteria should be taken

into consideration [7]. Bacteria can develop resistance against phages through various

mechanisms, including the modification of phage receptor-encoding genes, innate immune

systems (such as CRISPR-Cas), and the presence of prophages in the bacterial genome [47,


135

48]. Here, O1 serotype strains B1, B4, and B10 were associated with natural phage

resistance/high levels of bacterial growth, whereas the strains with serotype O2 and serogroup

O78:H4 were associated with phage susceptibility/bacterial killing. Moreover, a ll phage-host

combinations including P10-18 were found in Group 1 (fully resistant combinations). These

phages would be excluded as candidates for phage therapy targeting the selected APEC strains.

P1 was the only phage not included in any Group 1 combinations and showed the greatest

bacterial killing potential. Accordingly, phages only included in Group 2, characterised by high

bacterial killing, are considered the most promising candidates for phage therapy. Multiple

phage-host-MOI combinations were characterised by initial bacterial killing followed by

bacterial growth (subgroups 3.1, 3.2, and 3.3), suggesting the emergence of phage-resistant

bacterial variants. Whether phages found in these subgroups should be considered suitable for

phage therapy depends on their specific applications and further studies are needed to

determine if the initial inhibition of bacterial growth for ~7 hours is sufficient to clear out the

infection.

Although in vitro experiments do not capture many in vivo realities, such experiments can give

significant insights into the phage-host dynamics and lead to interesting predictions, which

could be useful in phage therapy and exploited in appropriately designed in vivo models [13].

Recently, a framework (Clinical Phage Microbiology) with recommendations for in vitro

identification and the evaluation of phages intended for treatment was published [42]. One step

of the framework pipeline includes determination of the growth kinetics of liquid cultures and

highlights the need for a standardised quantitative assessment with reproducible scoring . The

methodology applied here constitutes such an assessment and may help to improve the

standardisation of the quantitative evaluation of phage candidates.

In conclusion, our methodology assessing the host-phage interaction in vitro provides a high-

throughput method for classifying bacterial growth dynamics in the presence of virulent

coliphages using measurements of bacterial growth by OD as inputs. The established in vitro

model was not only used to gain a better understanding of the phage PK/PD but can also be

applied as a screening method for selecting new suitable phage candidates for therapeutic

applications against pathogenic target bacteria. However, to fully understand the complexity

of these phage-host dynamics, the underlying mechanisms behind these different interactions

need to be deciphered.


136

References



virology-100114-054952.














[8] Pirnay, J.; Verbeken, G.; Rose, T.; Jennes, S.; Zizi, M.; Huys, I.; Lavigne, R.; Merabishvili, M.;

Vaneechoutte, M.; Buckling, A.; et al. Introducing Yesterday’s Phage Therapy in Today’s Medicine.

Future Virol., 2012, 7 (4), 379–390. https://doi.org/10.2217/fvl.12.24.

[9] Gordillo Altamirano, F. L.; Barr, J. J. Phage Therapy in the Postantibiotic Era. Clin. Microbiol. Rev.,

2019, 32 (2), 1–25. https://doi.org/10.1128/CMR.00066-18.

[10] Lu, T. K.; Koeris, M. S. The next Generation of Bacteriophage Therapy. Curr. Opin. Microbiol.,

2011, 14 (5), 524–531. https://doi.org/10.1016/j.mib.2011.07.028.










[14] Gill, J.; Hyman, P. Phage Choice, Isolation, and Preparation for Phage Therapy. Curr. Pharm.

Biotechnol., 2010, 11 (1), 2–14. https://doi.org/10.2174/138920110790725311.

[15] Storms, Z. J.; Sauvageau, D. Modeling Tailed Bacteriophage Adsorption: Insight into Mechanisms.

Virology, 2015, 485, 355–362. https://doi.org/10.1016/j.virol.2015.08.007.




[17] Cairns, B. J.; Timms, A. R.; Jansen, V. A. A.; Connerton, I. F.; Payne, R. J. H. Quantitative Models

of in Vitro Bacteriophage-Host Dynamics and Their Application to Phage Therapy. PLoS Pathog.,

2009, 5 (1), e1000253. https://doi.org/10.1371/journal.ppat.1000253.




137


[19] Broussou, D. C.; Toutain, P. L.; Woehrlé, F.; Garch, F. El; Bousquet -Melou, A.; Ferran, A. A.

Comparison of in Vitro Static and Dynamic Assays to Evaluate the Efficacy of an Antimicrobial

Drug Combination against Staphylococcus Aureus. PLoS One, 2019, 14 (1), e0211214.


[20] He, S.; He, H.; Chen, Y.; Chen, Y.; Wang, W.; Yu, D. In Vitro and in Vivo Analysis of Antimicrobial

Agents Alone and in Combination against Multi-Drug Resistant Acinetobacter Baumannii. Front.








[23] Nhung, N. T.; Chansiripornchai, N.; Carrique-Mas, J. J. Antimicrobial Resistance in Bacterial

Poultry Pathogens: A Review. Front. Vet. Sci., 2017, 4, 126.

https://doi.org/10.3389/fvets.2017.00126.

[24] Dziva, F.; Stevens, M. P. Colibacillosis in Poultry: Unravelling the Molecular Basis of Virulence of

Avian Pathogenic Escherichia Coli in Their Natural Hosts. Avian Pathol., 2008, 37 (4), 355–366.

https://doi.org/10.1080/03079450802216652.

[25] Bankevich, A.; Nurk, S.; Antipov, D.; Gurevich, A. A.; Dvorkin, M.; Kulikov, A. S.; Lesin, V. M.;

Nikolenko, S. I.; Pham, S.; Prjibelski, A. D.; et al. SPAdes: A New Genome Assembly Algorithm

and Its Applications to Single-Cell Sequencing. J. Comput. Biol., 2012, 19 (5), 455–477.

https://doi.org/10.1089/cmb.2012.0021.







[28] Alcock, B. P.; Raphenya, A. R.; Lau, T. T. Y.; Tsang, K. K.; Bouchard, M.; Edalatmand, A.; Huynh,

W.; Nguyen, A.-L. V; Cheng, A. A.; Liu, S.; et al. CARD 2020: Antibiotic Resistome Surveillance

with the Comprehensive Antibiotic Resistance Database. Nucleic Acids Res., 2019, 48 (D1), D517–

D525. https://doi.org/10.1093/nar/gkz935.

[29] Arndt, D.; Grant, J. R.; Marcu, A.; Sajed, T.; Pon, A.; Liang, Y.; Wishart, D. S. PHASTER: A Better,

Faster Version of the PHAST Phage Search Tool. Nucleic Acids Res., 2016, 44, W16–W21.

https://doi.org/10.1093/nar/gkw387.

[30] Overbeek, R.; Olson, R.; Pusch, G. D.; Olsen, G. J.; Davis, J. J.; Disz, T.; Edwards, R. A.; Gerdes,

S.; Parrello, B.; Shukla, M.; et al. The SEED and the Rapid Annotation of Microbial Genomes Using

Subsystems Technology (RAST). Nucleic Acids Res., 2014, 42 (D1), 206–214.

https://doi.org/10.1093/nar/gkt1226.

[31] Sørensen, P. E.; Van Den Broeck, W.; Kiil, K.; Jasinskyte, D.; Moodley, A.; Garmyn, A.; Ingmer,

H.; Butaye, P. New Insights into the Biodiversity of Coliphages in the Intestine of Poultry. Sci. Rep.,

2020, 10, 15220. https://doi.org/10.1038/s41598-020-72177-2.

[32] Konopacki, M.; Grygorcewicz, B.; Dołęgowska, B.; Kordas, M.; Rakoczy, R. PhageScore: A Simple

Method for Comparative Evaluation of Bacteriophages Lytic Activity. Biochem. Eng. J., 2020, 161,

107652. https://doi.org/10.1016/j.bej.2020.107652.


138

[33] R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for

Statistical Computing: Vienna, Austria 2020.



[35] Gayder, S.; Parcey, M.; Nesbitt, D.; Castle, A. J.; Svircev, A. M. Population Dynamics between

Erwinia Amylovora, Pantoea Agglomerans and Bacteriophages: Exploiting Synergy and

Competition to Improve Phage Cockta il Efficacy. Microorganisms, 2020, 8 (9), 1449.

https://doi.org/10.3390/microorganisms8091449.

[36] Xie, Y.; Wahab, L.; Gill, J. J. Development and Validation of a Microtiter Plate-Based Assay for

Determination of Bacteriophage Host Range and Virulence. Viruses, 2018, 10 (4).

https://doi.org/10.3390/v10040189.

[37] Qadir, M. I.; Mobeen, T.; Masood, A. Phage Therapy: Progress in Pharmacokinetics. Brazilian J.

Pharm. Sci., 2018, 54 (1), e17093. https://doi.org/10.1590/s2175-97902018000117093.

[38] Dąbrowska, K. Phage Therapy: What Factors Shape Phage Pharmacokinetics and Bioavailability?

Systematic and Critical Review. Med. Res. Rev., 2019, 39 (5), 2000–2025.

https://doi.org/10.1002/med.21572.

[39] Benala, M.; Vaiyapuri, M.; Visnuvinayagam, S.; George, J. C.; Raveendran, K.; George, I.;

Mothadaka, M. P.; Badireddy, M. R. A Revisited Two-Step Microtiter Plate Assay: Optimization of

in Vitro Multiplicity of Infection (MOI) for Coliphage and Vibriophage. J. Virol. Methods, 2021,

294, 114177. https://doi.org/10.1016/j.jviromet.2021.114177.

[40] Rajnovic, D.; Muñoz-Berbel, X.; Mas, J. Fast Phage Detection and Quantification: An Optical

Density-Based Approach. PLoS One, 2019, 14 (5), e0216292.


[41] Haines, M. E. K.; Hodges, F. E.; Nale, J. Y.; Mahony, J.; van Sinderen, D.; Kaczorowska, J.;

Alrashid, B.; Akter, M.; Brown, N.; Sauvageau, D.; et al. Analysis of Selection Methods to Develop

Novel Phage Therapy Cocktails Against Antimicrobial Resistant Clinical Isolates of Bacteria. Front.


[42] Gelman, D.; Yerushalmy, O.; Alkalay-Oren, S.; Rakov, C.; Ben-Porat, S.; Khalifa, L.; Adler, K.;

Abdalrhman, M.; Coppenhagen-Glazer, S.; Aslam, S.; et al. Clinical Phage Microbiology: A

Suggested Framework and Recommendations for the in-Vitro Matching Steps of Phage Therapy.

The Lancet Microbe, 2021, 2 (10), e555–e563. https://doi.org/10.1016/S2666-5247(21)00127-0.

[43] Monferrer, E.; Domingo-Calap, P. Virus-Host Coevolution as a Tool for Controlling Bacterial

Resistance to Phage Therapy. J. Biotechnol. Biomed., 2019, 2 (3), 96–104.

https://doi.org/10.26502/jbb.2642-91280013.

[44] Lindberg, H. M.; McKean, K. A.; Wang, I.-N. Phage Fitness May Help Predict Phage Therapy

Efficacy. Bacteriophage, 2014, 4 (4), e964081. https://doi.org/10.4161/21597073.2014.964081.




[46] Koonjan, S.; Seijsing, F.; Cooper, C. J.; Nilsson, A. S. Infection Kinetics and Phylogenetic Analysis

of VB_EcoD_SU57, a Virulent T1-Like Drexlerviridae Coliphage. Front. Microbiol., 2020, 11,

565556. https://doi.org/10.3389/fmicb.2020.565556.



[48] Bondy-Denomy, J.; Qian, J.; Westra, E. R.; Buckling, A.; Guttman, D. S.; Davidson, A. R.; Maxwell,

K. L. Prophages Mediate Defense a gainst Phage Infection through Diverse Mechanisms. ISME J.,

2016, 10 (12), 2854–2866. https://doi.org/10.1038/ismej.2016.79.Güemes, A. G. C.; Merry Youle;


139

Cantú, V. A.; Felts, B.; Nulton, J.; Rohwer, F. Viruses as Winners in the Game of Life. Annu. Rev.

Virol., 2016, 3 (1), 197–214. https://doi.org/10.1146/annurev-virology-100114-054952.


140

Su

pp

lem

en

tary F

igu

re

S1

| G

row

th d

ynam

ics

cu

rves

for

phage-b

acte

rium

co

mbin

ati

ons

at

all M

OIs

. P

hage-f

ree c

ontr

ol cu

lture

s fo

r ea

ch A

PE

C s

train

are

inclu

ded

(E

mpty

). B

acte

rial g

row

th (O

D6

00) w

as

mea

sure

d f

or 2

2 h

.


141

Supplementary Figure S2 | Principal component analysis (PCA) of the differences in OD trajectories/values of

the phage-host combinations. The analysis captures 97% of the variance and grouped the total of 2869 entities in

three groups. Group 1 comprises resistant combinations with bacterial growth. Group 2 comprises susceptible

combinations with bacterial killing. Group 3 comprises in-between combination.


142

Supplementary Figure S3 | Growth dynamics curves for Group 3 subgroups. The subgroups included Group 3.1

(purple), Group 3.2 (orange), Group 3.3 (yellow), Group 3.4 (brown), and Group 3.5 (pink). Bacterial growth

(OD) was measured for 22 hours at 600 nm (OD600).


143

Supplementary Table S1 | Prophages identified in the APEC genomes

Prophage hit GenBank Length (Kb)

Completeness Score # Total Proteins

APEC Strain

Sero-group

Entero_BP_4795 NC_004813 6 incomplete 40 7 B2 O78:H4

Entero_BP_4795 NC_004813 8.2 incomplete 20 16 B8 O1:H7

Entero_cdtI NC_009514 31.1 incomplete 60 28 B3 O78:H4

Entero_cdtI NC_009514 6.7 incomplete 30 7 B9 O78:H9

Entero_DE3 NC_042057 37 incomplete 60 46 B10 O2:H5

Entero_DE3 NC_042057 12.1 incomplete 60 15 B10 O2:H5




Entero_fiAA91_ss NC_022750 25.9 intact 150 32 B2 O78:H4

Entero_HK629 NC_019711 17.1 questionable 70 27 B5 O2:H5

Entero_HK629 NC_019711 8.5 incomplete 20 11 B2 O78:H4

Entero_HK630 NC_019723 15.3 incomplete 20 18 B3 O78:H4

Entero_HK630 NC_019723 13 incomplete 60 16 B2 O78:H4

Entero_JenP1 NC_029028 5.3 incomplete 30 6 B1 O1:H7

Entero_lambda NC_001416 28.2 incomplete 10 26 B2 O78:H4




Entero_lambda NC_001416 10.8 questionable 70 15 B1 O1:H7

Entero_mEp460 NC_019716 14 incomplete 40 19 B6 O78:H9


Entero_mEp460 NC_019716 28.1 intact 150 41 B3 O78:H4


Entero_mEp460 NC_019716 21 intact 150 29 B1 O1:H7



Entero_mEp460 NC_019716 6.8 incomplete 50 17 B4 O1:H7









Entero_N15 NC_001901 20.2 incomplete 20 31 B7 O78:H9

Entero_Sf101 NC_027398 38.7 intact 120 61 B1 O1:H7

Entero_SfI NC_027339 9.8 incomplete 30 18 B4 O1:H7

Entero_SfI NC_027339 9.4 incomplete 20 17 B8 O1:H7

Entero_SfI NC_027339 32 incomplete 30 31 B1 O1:H7

Entero_SfV NC_003444 14.1 incomplete 40 9 B9 O78:H9

Entero_SfV NC_003444 10.5 incomplete 60 20 B7 O78:H9

Entero_Wphi NC_005056 23.3 intact 110 28 B7 O78:H9

Entero_YYZ_2008 NC_011356 10.4 incomplete 30 12 B2 O78:H4

Erwini_PEp14 NC_016767 9.1 incomplete 40 14 B9 O78:H9

Escher_500465_1 NC_049342 18.5 incomplete 60 12 B6 O78:H9



Escher_500465_1 NC_049342 23.8 questionable 86 29 B1 O1:H7



144

Supplementary Table S1 | Prophages identified in the APEC genomes

Prophage hit GenBank Length

(Kb) Completeness

Scor

e

# Total

Proteins

APEC

Strain

Sero-

group

Escher_Stx1 NC_004913 13 incomplete 60 14 B3 O78:H4

Escher_phiV10 NC_007804 45.5 intact 98 43 B2 O78:H4

Escher_phiV10 NC_007804 48.5 intact 100 45 B3 O78:H4

Escher_pro483 NC_028943 31.2 intact 150 32 B1 O1:H7

Escher_RCS47 NC_042128 54 incomplete 50 52 B9 O78:H9

Escher_SH2026Stx1 NC_049919 11.1 questionable 70 18 B2 O78:H4

Escher_SH2026Stx1 NC_049919 7 incomplete 40 11 B3 O78:H4

Escher_SH2026Stx1 NC_049919 5 questionable 70 10 B4 O1:H7

Escher_SH2026Stx1 NC_049919 5 incomplete 60 9 B8 O1:H7

Escher_SH2026Stx1 NC_049919 14.4 incomplete 10 18 B5 O2:H5

Flavob_FCL_2 NC_027125 7 incomplete 50 10 B1 O1:H7

Klebsi_4LV2017 NC_047818 15.4 incomplete 10 24 B9 O78:H9

Klebsi_ST437_

OXA245phi4.2 NC_049449 22.2 incomplete 10 13 B6 O78:H9

Klebsi_ST437_


Klebsi_ST437_


Marino_P12026 NC_018269 9.1 incomplete 40 14 B1 O1:H7

Mycoba_Gaia NC_026590 7 incomplete 50 11 B4 O1:H7

Pectob_ZF40 NC_019522 29.8 incomplete 40 25 B1 O1:H7



Pseudo_phiPSA1 NC_024365 20 intact 100 25 B3 O78:H4

Rhodoc_RGL3 NC_016650 9.1 incomplete 40 15 B7 O78:H9

Salmon_118970_sal

3 NC_031940 24.5 intact 150 36 B3 O78:H4

Salmon_118970_sal

3 NC_031940 14.9 incomplete 30 28 B7 O78:H9

Salmon_118970_sal

3 NC_031940 8.9 incomplete 60 14 B7 O78:H9

Salmon_SJ46 NC_031129 45 questionable 70 68 B9 O78:H9

Salmon_SJ46 NC_031129 7.6 incomplete 10 6 B6 O78:H9

Salmon_SSU5 NC_018843 9.1 incomplete 40 13 B2 O78:H4

Shigel_POCJ13 NC_025434 24.4 incomplete 20 21 B3 O78:H4

Shigel_SfII NC_021857 26.2 intact 150 37 B1 O1:H7

Shigel_SfII NC_021857 10.2 incomplete 60 13 B6 O78:H9

Staphy_SPbeta_like NC_029119 17.5 incomplete 40 8 B3 O78:H4

Stx2_c_1717 NC_011357 3.4 incomplete 60 7 B10 O2:H5

Stx2_c_Stx2a_F451 NC_049924 6.7 incomplete 60 10 B3 O78:H4

Vibrio_X29 NC_024369 26.7 intact 150 32 B2 O78:H4

Vibrio_X29 NC_024369 30 intact 150 34 B7 O78:H9

Yersin_L_413C NC_004745 23.3 intact 110 28 B6 O78:H9

Yersin_L_413C NC_004745 27.9 intact 120 29 B4 O1:H7

Yersin_L_413C NC_004745 7.3 incomplete 30 7 B8 O1:H7

Supplementary Table S2: Non-metric multidimensional scaling (NMDS) grouping summary and

Supplementary Table S3: Non-metric multidimensional scaling (NMDS) Group 3 subgrouping

summary are not included in this thesis due to their large size. Both tables are available upon request.


145

3.3

Spontaneous phage resistance in avian pathogenic Escherichia

coli

Patricia E. Sørensen1,2, Sharmin Baig3, Marc Stegger3, Hanne Ingmer5, An Garmyn1,

and Patrick Butaye1,2

1 Department of Pathology, Bacteriology and Poultry diseases, Ghent University, Belgium


Indies

3 Department of Bacteria, Parasites and Fungi, Statens Serum Institut, Copenhagen, Denmark


Published in Frontiers in Microbiology 2021, 12 (782757)



146

Abstract

Avian pathogenic Escherichia coli (APEC) is one of the most important bacterial pathogens

affecting poultry worldwide. The emergence of multidrug-resistant pathogens has renewed the

interest in the therapeutic use of bacteriophages (phages). However, a major concern for the

successful implementation of phage therapy is the emergence of phage-resistant strains. The

understanding of the phage-host interactions, as well as underlying mechanisms of resistance,

have shown to be essential for the development of a successful phage therapy. Here, we

demonstrate that the strictly lytic Escherichia phage vB_EcoM-P10 rapidly selected for

resistance in the APEC ST95 O1 strain AM621. Whole-genome sequence analysis of 109

spontaneous phage-resistant mutant strains revealed 41 mutants with single-nucleotide

polymorphisms (SNPs) in their core genome. In 32 of these, a single SNP was detected while

two SNPs were identified in a total of nine strains. In total, 34 unique SNPs were detected. In

42 strains, including 18 strains with SNP(s), gene losses spanning 17 different genes were

detected. Affected by genetic changes were genes known to be involved in phage resistance

(outer membrane protein A, lipopolysaccharide-, O-antigen-, or cell wall-related genes) as well

as genes not previously linked to phage resistance, including two hypothetical genes. In several

strains, we did not detect any genetic changes. Infecting phages were not able to overcome the

phage resistance in host strains. However, interestingly the initial infection was shown to have

a great fitness cost for several mutant strains, with up to ~65% decrease in overall growth. In

conclusion, this study provides valuable insights into the phage-host interaction and phage

resistance in APEC. Although acquired resistance to phages is frequently observed in

pathogenic E. coli, it may be associated with loss of fitness, which could be exploited in phage

therapy.


147

Introduction

Bacteriophages (phages) are viruses that specifically infect bacteria, and are estimated to be

the most abundant organisms on Earth with more than 1031 entities [1]. Phages are unable to

replicate independently of a susceptible bacterial host, and their host range is determined by a

combination of various factors, including specificity of host-binding phage proteins and

bacterial phage-resistance mechanisms [2, 3]. Virulent phages are strict parasites of their host

and confer a selective pressure on their host population through host cell lysis [4]. In response,

bacteria can evolve resistance to phage infection through various mechanisms, such as

spontaneous mutations, acquisition of restriction-modification (R-M) systems, and adaptive

immunity via Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)-Cas

system(s). These mechanisms can be used to target dif ferent steps of the phage life cycle,

including phage attachment, adsorption, replication, and host cell lysis [5, 6]. The different

resistance mechanisms result in distinct resistance phenotypes. These can differ in whether the

resistance is partial or complete, in the fitness cost associated with resistance, and in whether

the mutation can be countered by a mutation in the infecting phage [7, 8]. Although various

antiviral defense systems are found in bacteria, the emergence of phage resistance as well as

phage-bacterium co-evolution are often driven by spontaneous mutations [6, 9], which may

confer phage resistance by modifying phage-associated receptors on the bacterial surface.

However, such changes have also been associated with reduced fitness relative to non-resistant

strains [10]. Phage-resistant bacteria may become less virulent as in case when mutations occur

in their lipopolysaccharides (LPS), or may experience impaired growth in case of mutations in

genes involved in essential cell functions [11]. Additionally, maintenance of defense systems

such as R-M enzymes and CRISPR-Cas, also has its own costs associated with enzyme

production and expression [12–14].

Avian pathogenic Escherichia coli (APEC) is one of the most important bacterial pathogens

affecting poultry. These pathogens cause a large range of extra-intestinal infections, which

collectively are referred to as colibacillosis. These infections can result in high morbidity and

mortality, and hereby, significant economic loses to the poultry industry worldwide [15–18].

Here, the APEC with O-serogroups O1, O2 and O78 constitute more than 80% of the infection

cases [19]. As current antimicrobials become increasingly inadequate to treat bacterial

infections and a global focus to reduce conventional antimicrobial usage in general, alternative

treatment strategies, such as the therapeutic use of phages (phage therapy), are urgently needed

[20–22]. However, being able to understand phage-host interactions as well as the underlying


148

mechanisms of resistance is essential for successful phage therapy application [23]. Here, we

investigate the phage-host interactions and resistance through isolation and characterisation of

spontaneous phage-resistant mutants of APEC.

Materials and methods

Bacterial strains and growth conditions

The APEC ST95 O1:H7 strain, AM621, is part of the in-house collection that was isolated from

clinical material suspected of APEC infection from Belgium collected during 2013-2014 by

Animal Health Care Flanders (Torhout, Belgium). The E. coli K-12 derived laboratory strain

K514 [24] was included as a phage-susceptible control and host strain. Bacterial strains were

grown in LB broth (Miller) or on LB agar supplemented with 1.5% bacteriological agar no. 1

(w/v) (Oxoid, Thermo Fisher Scientific, USA) overnight (16-18 h) at 37°C unless stated

otherwise. Broth cultures were incubated with shaking (120 rpm). Strains were stored at -80°C

in LB broth supplemented with 15% glycerol.

Bacteriophage isolation and propagation

The strictly virulent Escherichia phage vB_EcoM-P10 (SRA accession no. SRX8360061) used

in this study is a part of the in-house phage collection (Phage P1 from Chapter 3.2). The phage

was isolated from poultry feces and processed as previously described [25]. Phage lysates were

stored at 4°C, at titers ranging from ~1.2 x 108 to 1.4 x 109 PFU/ml. Escherichia phage

vB_EcoM-P10 was classified (according to the International Committee on Taxonomy of

Viruses (ICTV) taxonomy) as a tailed Myoviridae phage belonging to the Tevenvirinae

subfamily and Tequatrovirus genus.

Isolation of phage-resistant mutant strains

Phage-resistant APEC strains were obtained using the agar plate (AP) [26] and the secondary

culture (SC) technique [27] with minor modifications (Supplementary Figure S1). Briefly,

overnight culture of WT strain AM621 was inoculated in LB broth supplemented with CaCl2

(final concentration of 10 mM) and then infected with suspension of virulent phage vB_EcoM-

P10, at a multiplicity of infection (MOI) of 0.1, 1, 10, and 100. For the AP technique,

suspensions were streaked directly onto LB agar plates supplemented with CaCl2 (final


149

concentration of 10 mM) and incubated for 48 h at 37°C. After incubation of 24 h and 48 h,

individual colonies were selected from each MOI suspension and cultured in LB broth. Isolates

were purified by three consecutive streakings on LB agar (in the absence of phage) and

recovered as presumptive phage-resistant mutants. Remaining MOI cultures that were not

streaked on agar plates were subjected to the SC technique. Cultures were incubated at 37°C

with shaking (120 rpm) for ~5 hours. Cultures exhibiting complete or partial lysis and

subsequent (secondary) growth after an additional incubation of 24 h were selected and

streaked on LB agar plates. Remaining “SC-T24” solutions were stored at 4°C until required.

Presumptive phage-resistant mutants were recovered as described for the AP technique and

stored at 4°C until required. An experiment with phage-susceptible E. coli laboratory strain

K514 was performed in parallel as control. The AP/SC experiments were repeated six times.

Presumptive phage-resistant mutants were infected with phage vB_EcoM-P10 using the fitness

test experimental set-up (described below). Mutants that displayed normal bacterial growth or

increased growth compared to the phage-sensitive AM621 WT strain were defined as true

phage-resistant mutants and stored at -80°C in LB broth supplemented with 15% glycerol (v/v).

Efficiency of the phage-resistant mutant recovery was calculated according to the formula

presented by Capra et al. (2011): (number of true phage-resistant mutants ⁄ number of

presumptive phage-resistant mutants) * 100.

Isolation and enumeration of potential phage mutants

To isolate potential phage mutants, the SC-T24 solutions were centrifuged and filtered using a

0.2 µm filter (Whatman, GE Healthcare, Germany). The filtrated SC-T24-phage suspensions

were enumerated and tested for lytic activity on the host bacteria, E. coli K-12 derived

laboratory strain K514, using the double-layer agar (DLA) technique [29]. Briefly, phage

suspensions were serial diluted and spotted on an overlay of the host bacteria on LB agar

supplemented with 0.7% agar and 0.5 mM CaCl2. A clear zone in the plate, a plaque, resulting

from the lysis of host bacterial cells, indicated the presence of virulent phage. Phage lysates

were stored at 4°C until required.

Bacterial fitness

Bacterial reduction experiments were performed as described previously [30, 31], with minor

modifications. Bacterial overnight cultures were used, and the cell concentration was adjusted


150

to ~108 CFU/ml for every experiment. Bacterial suspensions were inoculated with phage,

yielding MOIs of 0.1, 1, 10, and 100. All bacterial reduction curves were generated using 96-

well plates with working volumes of 200 µl. The experiment was carried out in duplicates and

repeated three times. Two wells of phage-free bacterial cultures and two wells of bacteria-free

phage culture were included on every plate as control experiments in addition to one media

blank for reference. Optical density (OD) for the wavelength of 600 nm was measured with the

Thermo Fisher Scientific Multiskan GO Microplate Spectrophotometer and the data were

recorded using the SkanIt Software, v6.0.2.3. OD600 measurements were taken immediately

after inoculation and then at 30 minutes intervals afterwards for 22 h. The protocol parameters

included incubation temperature of 37°C and continuous shaking with medium speed.

Reduction curves were obtained by plotting OD600 values after baseline adjustments against

time. For each reduction curve, area under the curve (AUC) was calculated using GraphPad

Prism v9.1.0.221 with default settings. AUC was calculated as average of four replicates.

Strains were defined as truly resistant when % of decrease in AUC in the presence of phage

was minimum 20% less relative to the WT strain. Fitness cost associated with acquired

mutations in true resistant strains was defined as decrease in AUC compared to WT strain in

the absence of phage.


Genomic DNA was extracted from true phage-resistant bacterial strains using Qiagen’s

DNeasy Blood and Tissue Kit (Qiagen, Hilden, Germany), with subsequent library

construction using the Nextera XT Kit (Illumina, Little Chesterford, UK) using a 300-cycle kit

on the Illumina NextSeq 550 platform according to the manufacturer’s instructions.

Phage DNA was extracted and purified using Phage DNA Isolation Kit (Norgen Biotek Corp.,

Canada), as indicated by the instructions provided by the manufacturer. The DNA yield was

quantified using the QuantiFluor dsDNA System (Promega) and Quantus Fluorometer. The

DNA purity (OD 260/280 ratio of ~1.7-1.8) was measured using NanoDrop (Isogen Life

Science). Libraries were constructed using the Nextera XT Kit (Illumina, Little Chesterford,

UK) using a 300-cycle kit on the Illumina NextSeq platform according to the manufacturer’s

instructions.


151


The open-source bifrost software (https://github.com/ssi-dk/bifrost), v1.1.0, was used for

quality control of the WGS data. The raw reads were de novo assembled using SPAdes v3.11.1

[32], and contigs with less than 200 bp were excluded. APEC serotype was predicted for each

of the strains using SerotypeFinder, v2.0 [33]. Genomes were annotated using Prokka, v1.12

[34], and pan genome analysis was carried out with Roary, v.3.12.0 [35], with minimum 90%

similarity on protein level. Gene presence was subsequent confirmed using Mykrobe predictor,

v0.5.6 [36]. Genes classified as present were further filtered for coverage (c>70) and depth

(d>3). When inconsistencies were observed, manual BLASTn searches were performed. Cases

where a gene was detected in a mutant strain but not in the WT strain were excluded from

further analysis, as this was assumed to be sequencing error or contamination (a false-positive).

PlasmidFinder 2.1 with default settings was used to screen assembled genomes for plasmids in

the Enterobacteriaceae database. Plasmid replicons with less than 90% identity and 60%

coverage were excluded. ABRicate v1.0.1 (https://github.com/tseemann/abricate) with default

options was used to screen assembled genomes for antimicrobial resistance genes with

ResFinder database [37], NCBI Bacterial Antimicrobial Resistance Reference Gene Database

[38], and the Comprehensive Antibiotic Resistance Database (CARD) [39]. Virulence genes

were identified using ABRicate with sequences from the Ecoli_VF database. CRISPR systems

were identified using the Geneious Prime v2020.1.1 Crispr Recognition Tool Wrapper v1.1.

and CRISPRCasFinder (https://crisprcas.i2bc.paris-saclay.fr/CrisprCasFinder/Index) [40] with

default settings. A quality score was automatically given to CRISPR arrays consisting of

repeats and spacer sequences in the form of “evidence level”, rated 1-4, where 1 includes small

CRISPRs (with three or less spaces) and 2-4 are classified based on repeat and spacer similarity.

BLAST analysis was performed to determine if identified CRISPR spacer sequences matched

the invading Escherichia phage vB_EcoM-P10 genome.

Bacterial core genome SNP analysis

To assess the relationship between strains, a single nucleotide polymorphism (SNP)-based

phylogeny was obtained using SNPs identified by the Northern Arizona SNP Pipeline (NASP),

v1.2.0 [41], with the Burrows-Wheeler Aligner (BWA) algorithm, v0.7.17-r1188 [42].

Illumina reads from all individual strains were aligned against the AM621 WT scaffold genome

obtained as described with a cut-off of all contigs <500 bp above after removal of duplicated

https://github.com/ssi-dk/bifrost

https://github.com/tseemann/abricate

https://crisprcas.i2bc.paris-saclay.fr/CrisprCasFinder/Index


152

regions using NUCmer, v3.1 [43]. Positions with less than 10-fold coverage and less than 90%

unambiguous variant calls were excluded across the collection. The chromosome from the

well-characterised ST95 E. coli isolate UTI89 (GenBank accession number NC_007946) was

used to infer functionality of all the identified SNP differences.

Phage genome analysis

Phage genome analysis, including quality control validation, de novo assembly, annotation,

and pan genome analysis, was performed as described above for the bacterial genomes. Core

genome SNP analysis was performed as described for the bacterial genome using the

chromosome from the highly similar, well-characterised Escherichia phage vB_EcoM_G29

(GenBank accession number MK327940) as reference.

Results

Isolation of phage-resistant mutants

A total of 264 presumptive phage-resistant variants were obtained from the AP and SC methods

using the strictly virulent Myoviridae phage vB_EcoM-P10. Only 109 isolates (~41%) were

considered true phage-resistant derivatives based on increase in bacterial growth (area under

the curve (AUC)) relative to the WT strain in the presence of phage. In this study, the SC

method generated more mutants than the AP method (Table 1).

Table 1. Phage-resistant mutants isolated using secondary culture (SC) or agar plate (AP)

methods

No. of presumptive phage

resistant mutants

No. of true phage-resistant

mutants Isolation efficiency

AP 132 33 25%

SC 132 76 58%

For the AP method, the highest number of true resistant mutants were isolated from MOI 100

suspensions (~42%) and the lowest from MOI 0.1 (0%). For the SC method, the highest number

of true resistant mutants were isolated from MOI 1 suspensions (~30%) and the lowest from


153

MOI 100 (~21%). Similar numbers of true resistant mutants were isolated after 24 h and 48 h

of incubation (Supplementary Figure S2).

Bacterial fitness

The fitness cost associated with acquired mutation(s) in phage-resistant strains was determined

as decrease in overall bacterial growth (AUC) relative to the WT strain in the absence of phage

(Supplementary Table S1). The greatest fitness cost was detected for mutant strain SC48_10_8

(65% growth reduction), followed by AP48_1_24 (59%) and SC24_01_5 (57%). A fitness cost

of 31.6-37.5% was observed for five mutants. A fitness cost of 22.0-28.7% was observed for

four mutants. A fitness cost of 10.4-18.8% was observed for 24 mutants. A fitness cost of 5.2-

9.9% was observed for 33 mutants, and low or no fitness cost (<5%) was observed for 39 of

the mutant strains (Figure 1).

Figure 1. Decrease in growth of the phage-resistant APEC strains. The fitness cost associated with acquired

genetic changes in phage-resistant strains was determined as percentage decrease in overall bacterial growth (area

under the curve).


154


WGS of the bacterial genomes yielded a total of 1,934,298- 6,753,240 paired-end reads for

each isolate with an average coverage of 51-177-fold. De novo assembly resulted in 192-353

contigs and an N50 value from between 51,335- 189,445 bp.

The bacterial strains were subjected to WGS analysis. All 109 resistant strains showed similar

genetic characteristics as the AM621 WT, including a genome size between ~5.27 and ~5.40

Mbp and G+C content between 50.2 and 50.6%. Gene absence/presence analysis identified a

total of 17 different accessory genes (after exclusion of false positives), that were lost (partial

or complete) in one or more of mutant strains (Figure 2 and Table 2). A full overview of the

genes lost in phage-resistant mutants is shown in Supplementary Table S2. None of the mutant

strains lost any plasmid replicons compared to the WT. The six plasmid replicons detected

included Col(MG828), IncFIA, IncFIB(AP001918), IncFIC(FII), IncI1-I(Alpha), and IncX1.

Figure 2. Genes affected by genetic changes in phage-resistant APEC strains. In total, 44 different genes were

affected by genetic change(s). Genetic changes included complete gene loss, partial gene loss, or point mutations

(nonsense, missense, or synonymous). Full circle = complete gene loss in few mutants or nonsense mutation in

one mutant. Striped circle = missense mutation in one mutant. * = protein name is shown as gene name is

unknown.


155

Ta

ble

2 | S

um

mary

of

gen

eti

c c

hanges

an

d a

ffecte

d g

en

es

in p

hage

-res

ista

nt E

. co

li s

train

s

Ch

an

ge

Aff

ecte

d

gen

e

An

no

tati

on

F

un

cti

on

N

o. o

f

stra

ins

Ref(

s).

Gen

e lo

ss

neu

E

Poly

sial

ic a

cid b

iosy

nth

esis

pro

tein

, N

euE

E

. co

li K

1 s

iali

c ac

id c

apsu

le s

ynth

esis

24

[4

4]

Gen

e lo

ss

gro

up_67

H

ypo

thet

ical

pro

tein

U

nknow

n

24

-

Gen

e lo

ss

gro

up

_2

37

H

yp

oth

etic

al p

rote

in

Un

kn

ow

n

10

-

Gen

e lo

ss

gro

up

_3

10

T

etra

tric

op

epti

de

rep

eat (T

PR

) p

rote

in

Med

iati

on

of

pro

tein

-pro

tein

inte

ract

ions

8

[45

]

SN

P

ack

A

Ace

tate

kin

ase

Phosp

hory

lati

on o

f ac

etat

e to

ace

tyl phosp

hat

e 7

[46

]

SN

P

SN

P1

*

Hy

po

thet

ical

pro

tein

U

nk

no

wn

6

-

Gen

e lo

ss

mp

rA_

1

Mar

R f

amil

y t

ran

scri

pti

on

al r

egu

lato

r R

egu

lati

on

of

num

erous

cell

ula

r pro

cess

es

5

[47

]

Gen

e lo

ss

rfbX

O

-anti

gen

tra

nsp

ort

er

Tra

nsp

ort

of

O-p

oly

sacc

har

ide

mole

cule

s 4

[48

]

SN

P

gly

S

Gly

cyl-

tRN

A s

yn

thet

ase

bet

a ch

ain

tR

NA

rec

og

nit

ion

4

[49

]

Gen

e lo

ss

fad

E

Acy

l-C

oA

deh

yd

rog

enas

e D

ehy

dro

gen

atio

n o

f ac

yl-

coen

zym

es A

3

[50

]

Gen

e lo

ss

yafV

Y

afV

(2

-oxoglu

tara

mat

e am

idas

e)

Met

aboli

te r

epai

r en

zym

e 3

[51

]

Gen

e lo

ss

ivy

Ver

tebra

te lyso

som

e in

hib

itor

Pro

tect

ion

ag

ainst

lyso

zym

e-m

edia

ted c

ell w

all

hy

dro

lysi

s 3

[52

]

SN

P

gln

D

Pro

tein

-PII

uri

dyly

ltra

nsf

eras

e N

itro

gen

reg

ula

tion

3

[53

]

SN

P

pta

P

hosp

hat

e ac

etylt

ransf

eras

e A

ceta

te m

etab

oli

sm

3

[46

]

Gen

e lo

ss

epsM

_1

A

cety

ltra

nsf

eras

e / N

euD

pro

tein

E

. co

li K

1 s

iali

c ac

id c

apsu

le s

yn

thes

is

2

[54

]

Gen

e lo

ss

wzy

O

1 f

amil

y O

-anti

gen

poly

mer

ase

Synth

esis

of

the

LP

S B

-ban

d O

anti

gen

2

[55

]

SN

P

om

pA

O

ute

r m

embra

ne

pro

tein

A (

Om

pA

)

Key

E. co

li v

irule

nce

fac

tor

2

[56

, 5

7]

SN

P

relA

G

TP

py

rop

ho

sph

ok

inas

e S

yn

thes

is o

f p

pG

pp f

rom

GT

P

2

[58

]

SN

P

pyk

F

Pyru

vat

e kin

ase

Reg

ula

tion o

f th

e gly

coly

tic

pat

hw

ay

2

[59

]

Gen

e lo

ss

wek

M

Gly

cosy

ltra

nsf

eras

e fa

mil

y 4

P

epti

dogly

can b

iosy

nth

esis

1

[48

, 6

0]

Gen

e lo

ss

gro

up

_2

71

H

yp

oth

etic

al p

rote

in

Un

kn

ow

n

1

-

Gen

e lo

ss

hokA

H

okA

T

oxin

of

a ty

pe

I to

xin

-anti

toxin

(T

A)

syst

em

1

[61

]

Gen

e lo

ss

ydfO

D

UF

139

8 f

amil

y p

rote

in, Y

dfO

U

nknow

n

1

[62

]

Gen

e lo

ss

neu

A

Acy

lneu

ram

inat

e cy

tid

yly

ltra

nsf

eras

e E

. co

li K

1 s

iali

c ac

id c

apsu

le s

yn

thes

is

1

[54

]

Gen

e lo

ss

splE

S

erin

e pro

teas

e S

plE

In

volv

ed i

n v

ario

us

bio

logic

al p

roce

sses

1

[63

]

SN

P

dsb

B

Per

ipla

smic

thio

l:dis

ulf

ide

oxid

ore

duct

ase

Dsb

B

Ele

ctro

n tra

nsf

er c

atal

yst

1

[64

]

SN

P

wzz

B

O-a

nti

gen

ch

ain

len

gth

det

erm

inan

t p

rote

in W

zzB

L

ipo

po

lysa

cchar

ide

(LP

S)

bio

synth

esis

1

[65

]

SN

P

met

G

Met

hio

ny

l-tR

NA

synth

etas

e P

rote

in b

iosy

nth

esis

1

[66

]

SN

P

SN

P11*

tR

NA

-Val

-GA

C

Tra

nsf

er o

f am

ino a

cids

to t

he

riboso

me

1

[67

]

SN

P

na

rP

Nit

rate

/nit

rite

res

po

nse

reg

ula

tor

pro

tein

Nar

P

Gen

e ex

pre

ssio

n r

egula

tion

1

[68

]


156

Ta

ble

2 | C

on

tin

ued

Ch

an

ge

Aff

ecte

d

gen

e

An

no

tati

on

F

un

cti

on

N

o. o

f

stra

ins

Ref(

s).

SN

P

glt

X

Glu

tam

yl-

tRN

A s

ynth

etas

e P

rote

in b

iosy

nth

esis

1

[69

]

SN

P

hcp

T

6S

S c

om

pon

ent H

cp

Bac

teri

al inte

ract

ion w

ith h

ost

cel

ls

1

[70

]

SN

P

rap

Z

RN

ase

adap

ter

pro

tein

Rap

Z

Cel

l en

vel

op

e pre

curs

or

sensi

ng a

nd s

ignal

ing

1

[71

–7

3]

SN

P

rbsA

M

on

osa

cch

arid

e-tr

ansp

ort

ing

AT

Pas

e T

ran

sfer

of

solu

tes

acro

ss m

embra

nes

1

[74

]

SN

P

xylB

X

ylu

lose

kin

ase

Phosp

hory

lati

on o

f D

-xylu

lose

to D

-xylu

lose

5-

ph

osp

hat

e 1

[75

]

SN

P

wa

aC

L

ipo

po

lysa

cch

arid

e co

re h

epto

sylt

ran

sfer

ase

I L

ipo

po

lysa

cchar

ide

(LP

S)

bio

synth

esis

1

[60

]

SN

P

waaY

L

ipop

oly

sacc

har

ide

core

hep

tose

(II

) kin

ase

Rfa

Y

Lip

opoly

sacc

har

ide

(LP

S)

bio

synth

esis

1

[60

]

SN

P

fur

Fer

ric

up

tak

e re

gu

lati

on

pro

tein

FU

R

Tra

nsc

rip

tio

nal

reg

ula

tion o

f ir

on m

etab

oli

sm

1

[76

]

SN

P

SN

P3

1*

5

'-n

ucl

eoti

das

e H

yd

roly

sis

of

the

phosp

hat

e gro

up o

f 5′-

nucl

eoti

des

1

[77

]

SN

P

phnD

P

ho

sph

on

ate

AB

C tra

nsp

ort

er s

ub

stra

te-b

ind

ing

pro

tein

Phn

D

Phosp

honat

e upta

ke

and u

tili

sati

on p

athw

ay

1

[78

]

SN

P

ybjT

U

nch

arac

teri

sed

pro

tein

Yb

jT

E. co

li L

PS

bio

synth

esis

and o

ther

core

cel

l en

vel

op

e co

mp

onen

ts

1

[79

]

SN

P

rne

Rib

on

ucl

ease

E

RN

A p

roce

ssin

g a

nd m

RN

A d

egra

dat

ion

1

[80

]

SN

P

sgcR

sg

c re

gio

n tra

nsc

ripti

onal

reg

ula

tor

Tra

nsc

ripti

onal

reg

ula

tion

1

[81

]

SN

P

SN

P37*

H

ypo

thet

ical

pro

tein

U

nknow

n

1

-


157

All but one mutant strain encoded the same resistance genes as the WT strain. Only one mutant

strain, AP24_100_8, had lost qnrS1, a quinolone resistance gene. A total of 226 different

virulence genes were all identified in both the WT strain and all the mutants. Two different

type I-F CRISPR systems (evidence level 4) were detected in the AM621 WT strain. The first

system comprised seven repeat units of 20 bp and six CRISPR spacers, including five spacers

of 40 bp and one spacer of 41 bp. The second system comprised six repeat units of 28 bp and

five spacers of 32 bp. Moreover, two additional small CRISPR-like structures (evidence level

1); one with only two CRISPR repeats (44 bp) and one spacer (52 bp) and another with only

two repeats (36 bp) and one spacer (59 bp) were separately identified in the genome. The same

two CRISPR systems and two small CRISPR-like elements were found in all 109 mutant

strains. Additionally, between one and eight evidence level one CRISPR-like structures, which

were not in the WT strain, were detected in 102 of the mutants (Supplementary Table S3). Only

three mutant strains, AP24_10_14, AP48_1_24, and SC24_01_5, had acquired a CRISPR-like

element spacer of 53 bp that matched the invading phage genome.

Bacterial core genome SNP analysis

SNP analysis identified between 0-2 SNP difference(s) in the core genome between AM621

and the mutants. Of the 109 mutants, 66 showed no SNP differences, 33 mutants showed one

SNP difference and 10 mutants showed two SNP differences (Figure 3). A summary of SNPs

identified in the mutants is shown in Table 2 and Figure 2. A total of 37 unique SNPs were

identified, five of which resulted in a nonsense mutation, 21 in a missense mutation, six in a

synonymous mutation, and five of which were found in non-coding regions when analysed

against the annotation of the UTI89 genome. Nonsense mutations were found in five different

genes, including acetate kinase (ackA), outer membrane protein A (ompA), phosphate

acetyltransferase (pta), LPS core heptosyltransferase I (waaC), and LPS core heptose (II)

kinase (waaY) (Figure 2). Missense mutations were found in 19 different genes (Supplementary

Table S2).


158

Figure 3. Phage-resistant strains and their genetic changes. In total, 44 different genes were affected by genetic

change(s). Blue = SNP mutation. Black = partial and/or complete gene loss. * = protein name is shown as gene

name is unknown.


159

Impact of selection methods on mutations

Number and type of genetic changes (gene loss or SNP) in the phage-resistant mutant strains

was compared in relation to selection method (AP or SC), including the four different MOIs,

0.1, 1, 10, and 100 (Figure 4). The SC method produced the highest number of genetic changes.

No genetic changes were detected in resistant strains generated using the AP-MOI-0.1 selection

method. For all other selection methods, gene loss was the dominant type of genetic change,

with the only exception of AP-MOI-1 where both gene loss and SNP were detected once.

Figure 4. Type of bacterial genetic change detected for each method and multiplicity of infection (MOI).

The number and type of genetic changes including SNP (grey) and partial or complete gene loss (black) organised

based on method used. AP = agar plate. SC = secondary culture, at the four different MOIs: 0.1, 1, 10, and 100.

Phage genome analysis

To investigate if the 24 co-cultured SC-24 phages had evolved to overcome phage resistance

mechanisms in the mutant strains, these phages, as well as the WT Escherichia phage

vB_EcoM-P10, were subjected to WGS. The WGS of the phage genomes yielded a total of

942,276-2,338,994 paired-end reads for each isolate with an average coverage of 794-1998-

fold. De novo assembly resulted in 18-206 contigs and an N50 value ranging from 167,139-

167,243 bp. Pan-genome analysis of the 25 coliphages included 271 genes. All genes were


160

detected in all potential mutant phages using BLAST. SNP analysis identified no SNP

differences in the core genome between WT Escherichia phage vB_EcoM-P10 and the

potential phage mutants.

Discussion

In this study, we selected and characterised phage-resistant mutant strains of O1 APEC strain

AM621. Using a combined approach of the SC and AP method resulted in an overall mutant

isolation efficiency of ~41%. Previous studies using this approach found an isolation efficiency

of true resistant Lactobacillus paracasei isolates of 56% [82] and an average isolation

efficiency of 36.5% (ranging between 29.5-50%) of true resistant Lactobacillus delbrueckii

isolates [83]. We found an SC method isolation efficiency of 57.6%, while the AP method

efficiency was much lower (25.0%). The higher efficiency of the SC method has been reported

before though with similar, smaller or larger differences [82–84]. The lower AP efficiency

(especially at low MOIs) could be explained by a low selection pressure for phage resistance.

When comparing the specific isolation percentages, one must take into consideration the

differences in how “true resistance” was defined as well as the differences of the bacterial WT

strains used. In our study, the resistant mutants were quantitatively defined (increased AUC

relative to the WT strain in the presence of phage) whereas previous studies used a qualitative

approach (visual comparison of turbidity between phage-host co-cultures and control culture)

to define true resistance. As opposed to the qualitative approach, defining true resistant mutants

based on AUC provide high-throughput assessment based on fixed cut-off values, which can

easily be compared, and do not depend on experience and/or subjectivity of the observer.

However, one must be aware of the potential pitf alls related to the AUC as selection criterion.

If the AUC increase percentage cut-off value is too high, true resistant mutants may wrongfully

be excluded. If the cut-off value is too low, this approach could select both resistant mutants

and non-resistant strains.

Bacteria have been shown to evolve resistance to phage infection through mechanisms of

adsorption inhibition, including loss or modification of phage receptors [6, 7, 85]. There is a

great diversity reported in coliphage receptors, which include bacterial oute r membrane

proteins (OMPs), porins, capsule and LPS [56, 72, 86]. OMPs participate in outer membrane

functionality, including diffusion and transport mechanisms, cell shape as well as virulence

[87]. Also, the OmpA protein is a key virulence factor of pathogenic E. coli playing a role in


161

conjugation, adhesion, immune system evasion, resistance to environmental stress [57].

Therefore, mutation in such gene, while conferring resistance, may decrease bacterial adhesion

and immune system evasion, and hereby, the overall strain virulence in vivo/in situ. In addition,

phage resistance may also have a fitness cost [11]. In this study, we observed up to 65%

decrease in in vitro fitness (bacterial growth) in mutant strains that had acquired resistance

through genetic mutations and/or gene loss. However, such fitness cost may vary in in vivo/in

situ environments, as the magnitude has been shown to depend on the genetic basis of the

resistance as well as on the environmental context [88].

Recently, Maffei et al. (2021) investigated the coliphage-host interaction and identified phage

receptors. In accordance with previous findings, Myoviridae coliphages belonging to the

Tequatrovirus genus were found to use the OMP, Tsx (T6-like phages), FadL (T2-like phages),

OmpA, OmpC (T4-like phages), or OmpF as primary receptor. A recent study similarly

identified the OmpA protein as a Myoviridae coliphage receptor and reported that all phage-

resistant strains had acquired mutations in just two pathways, the LPS biosynthesis and the

OmpA expression [90]. LPS are known to play an essential role in the OMP folding and

placement in the cell wall [91]. Accordingly, loss or changes in the structure of LPS could

prevent OmpA from being properly positioned in the outer membrane, and thereby, making

the phage receptor unavailable. In our study, we detected SNPs in the ompA gene, encoding

the OmpA protein, suggesting this could act as receptor for phage vB_EcoM-P10. However,

further studies are needed to confirm if OmpA is the primary receptor as well as determine the

indirect effects on infection due to LPS changes.

While for some phages the absence of the primary receptor results in complete absence of

infection, other phages, including those utilizing several receptors, are still able to infect [89,

92, 93]. The specificity for the second receptor depends on the short tail fibers of which two

variants have been described to date [89]. The first variant (encoded by phages such as T2, T4,

and T6) targets the lipid A Kdo region deep in the LPS core, and a second variant targets the

upper part(s) of the LPS core, which requires an intact inner LPS core for infectivity. The

Myoviridae phage used in this study clusters with the latter group [25]. We found genetic

changes in the gene encoding glycosyltransferase required for the assembly of the LPS as well

as in the genes encoding LPS inner core heptose (II) kinase (waaY) and heptosyltransferase I

(waaC). Accordingly, as both waaY and waaC are essential for the LPS inner core, the

nonsense mutations detected in these genes will most likely have an effect on the infectivity of

an infecting phage. Either a direct effect as shown for phages utilising the LPS as a receptor


162

[94] or an indirect effect where waa mutation(s) interfere with the recognition of outer

membrane protein phage receptors [95]. At the same time, mutants with truncated LPS at the

inner core have been shown to have attenuated in vivo virulence and to be more sensitive to

antimicrobials [86, 90].

The O-antigen biosynthesis operon has been shown to play a major role in E. coli phage

resistance against Myoviridae phage T4 [48] and Demerecviridae (previous Siphoviridae)

phage T5 [56]. In accordance with these previous observations, we found genetic changes

(missense mutation, partial or complete gene loss) in four O-antigen operon genes encoding a

glycosyltransferase, the O-antigen polymerase (wzy), a chain length determinant protein (wzzB

gene) [65], and the O antigen flippase (wzx gene / rbfX gene), all of which could potentially

confer phage resistance. These findings could support the LPS as a potential binding site for

our phage.

In both Gram-negative and Gram-positive bacteria, the RNase adaptor protein RapZ plays a

central role in regulatory pathway of glucosamine-6-phosphate (GlcN6P), an early and

essential precursor in the synthesis of the bacterial cell envelope components, including

peptidoglycan, LPS and colanic acid [96]. Recent studies have demonstrated that phage

resistance in E. coli and Staphylococcus aureus can be acquired through mutation(s) in the

rapZ gene, encoding RapZ [71, 72, 97]. Zhou et al. (2021) reported that mutation in the rapZ

gene conferred E. coli phage resistance by inhibiting 93.5% phage adsorption. In this study, we

similarly detected a missense mutation in the rapZ gene supporting its involvement in phage

resistance against lytic Myoviridae coliphages. Moreover, in according with finding of Zhou et

al. (2021), no in vitro fitness cost (measured by bacterial growth) was associated with the

acquired resistance.

The polysaccharide capsule of pathogen E. coli K1 is an essential virulence factor and consist

of polymers of sialic acid (NeuNAc). The kps gene cluster encodes six proteins, NeuDBACES,

required for synthesis, activation, and polymerisation of NeuNAc [54, 98, 99]. In this study,

we detected partial gene loss of neuD (involved in the synthesis of sialic acid) [54], neuA

(synthase involved in activation the sugar prior to polymerisation) [54], and neuE (involved in

synthesis and export of NeuAc) [44]. The capsule is recognised as a receptor by some phages,

such as K-specific coliphages and the Myoviridae coliphage phi92, which have virion-

associated polysaccharide-degrading enzymes [100, 101]. Contrary, Scholl and colleagues

(2005) showed that the expression of the E. coli K1 capsule physically blocks infection by


163

phage T7, a phage that recognise LPS core as the primary receptor. Whether or not our

Myoviridae phage can utilise the capsule as receptor needs to be investigated further.

Nevertheless, as polysaccharide capsule is a key virulence factor, the interesting finding that

~23% of the phage resistant isolates have lost part of one of the neu genes could add to the

phage therapy potential of the infecting phage. Being as the infection could result in reduced

virulence as well as reduced competitiveness. Accordingly, (partial) loss of neuE may be

associated with great fitness cost as up to ~65% growth decrease was observed for the phage-

resistant mutant strains. However, in all affected strains two or more other genetic changes

were detected, strongly implying that further studies are needed to determine the exact effect

of neuE loss alone and in combination with the other affected genes.

Even though we were able to connect some of the genetic changes in the mutant strains to

known phage resistance mechanisms, most SNPs (n = 23) and gene losses (partial or complete)

(n = 11) were found in a gene not previously linked to phage resistance. Among others, these

genes encode acetate kinase (essential for bacterial growth) [46], Acyl-CoA dehydrogenase

(involved in the beta-oxidation cycle of fatty acid degradation) [50], the MarR family

transcriptional regulator (involved in numerous cellular processes, including stress responses,

virulence, and efflux of harmful chemicals and antimicrobials) [47], pyruvate kinases (essential

for the regulation of the glycolytic pathway) [59], a tetratricopeptide repeat (TPR) protein

(involved in various biological processes and mediates protein-protein interactions) [45],

uridylyltransferase (involved in nitrogen regulation) [53] as well as several hypothetical

proteins. We found loss of the gene or mutation in an acetate kinase, pyruvate kinase, TPR

protein and uridylyltransferase as the sole genetic change indicating that the phage -host

interaction might be more complex that previous thought. Interestingly, partial or complete loss

of one of two genes (group_67 and group_237) encoding hypothetical proteins was detected

in a great number of phage-resistant mutant strains, and as sole resistance mechanisms in some.

Loss of group_67 gene as sole resistance mechanism resulted in an average fitness cost (growth

reduction) of only 6.3%. Similarly, loss of the neuE gene as sole resistance mechanism resulted

in an average fitness cost of only ~3.9%. However, the greatest fitness cost was observed for

the mutant strain that had lost both the group_67 and neuE (65.2%) or both genes in

combination with a point mutation in the phnD gene (57.0%), indicating that a combination

loss of group_67 and neuE might have an additive effect on the fitness cost. The point mutation

in phnD was only observed in one mutant and only in combination with group_67 and neuE

gene loss. Only one mutant had lost the group_237 gene as sole resistance mechanisms and


164

suffered a great fitness cost of 59.1%. Moreover, an average fitness cost of 23.2% was observed

for the 10 mutant strains with group_237 gene loss, suggesting that while mutation in this gene

might confer phage resistance, the resistance comes with a cost for the host bacterium.

Furthermore, one mutant had lost genes encoding both group_237 and group_67 and suffered

a fitness cost (34.6%), supporting the essential role of group_237 and the potential additive

effect of group_67 gene loss. However, as additional genetic changes (potentially related to

phage resistance) were detected in most of both the group_67 and group_237 mutant strains.

We tried to decipher the potential function of the hypothetical proteins, using PANDA [103]

and LocTree3 [104], however we could not find any motifs that could give an indication (data

not shown). Also, the role of these proteins in E. coli phage resistance needs to be further

investigated.

A nonsense mutation was detected in the gene encoding the YbjT protein. This protein has

been shown to be physically tethered to the inner membrane of E. coli and part of the metabolic

pathway involved in the biogenesis of the bacterial cell envelope [105]. However, as this

genetic change was not the only one detected in the affected strain, its potential involvement

in phage resistance remains to be investigated. Finally, six different synonymous SNPs were

identified in this study. Although unlikely, these mutations may still play a role in phage

resistance as synonymous mutations can affect cellular processes such as translation efficiency

or mRNA structures, depending on the gene affected [106].

CRISPR-Cas systems are found among ~36% of bacteria and confer a sequence specific

adaptive immunity against invading foreign DNA, including phages [107]. Previous studies

have reported varying findings when it comes to phage resistance conferred by acquired

CRISPR spacer(s). As opposed to findings of Denes et al. (2015) where no CRISPR immunity

was observed in any of the spontaneous phage-resistant Listeria mutant strains, in most of the

phage-resistant Streptococcus mutant strains one or two CRISPR spacer(s) were acquired

[109]. In this study, we found three phage-resistant strains with a newly acquired CRISPR

spacer sequence that matched the invading phage genome. This spacer was found in a short

CRISPR array, only consisting of this one spacer (evidence level 1), which makes it difficult

to determine if this array is a false CRISPR-like element or a true CRISPR. However, the lack

of similar repeats in larger CRISPR arrays, associated cas genes and leader sequence upstream

of the CRISPR array, are indications that the detected CRISPR spacer in the three phage-

resistant mutant strains most likely is a false positive [40]. Moreover, two out of the three

mutant strains had acquired one or three genetic changes in addition to the CRISPR-like spacer


165

acquisition, including partial loss of the group_237 gene or partial loss of neuE, complete loss

of the group_67 gene and a silent point mutation in a hypothetical protein. As discussed earlier,

the partial and/or complete gene loss(es) are more likely to explain the resistance observed.

Phages can evolve to counteract bacterial antiviral mechanisms, such as inhibition of phage

adsorption, R-M systems, CRISPR-Cas systems and phage escape strategies [9, 110]. Such

adaptation can be conferred by point mutations in specific genes, such as receptor binding

proteins (RBPs) and/or tail fibers, genome rearrangement, and genetic exchange with other

viral or bacterial genomes to new traits [110]. Phage genes involved in host recognition are

among the fastest evolving phage genes due to the selection pressures conferred by the phage-

bacterium co-evolution [95, 110]. Meyer et al. (2012) showed that a lytic coliphage was able

to evolve as such that it could use an alternative receptor after eight days of co-culture with a

resistant bacterial host. Similarly, Wandro et al. (2019) showed that after eight days of co-

culture the lytic Enterococcus Phage EfV12-phi1was able to combat phage-resistance through

adaptation of the tail fiber. Hall et al. (2011) were able to detect adaptation in Pseudomonas

phage SBW25Φ2 tail fiber protein and structural protein after only two and four days of co -

culture, respectively. As opposed to these findings, in this study we did not detect any genetic

changes in phages co-cultured with phage-resistant strains. However, this is most likely a

reflection of a too short co-culture incubation period (<24 hours) rather than the ability of the

phage to co-evolve to bypass the phage resistance.

Understanding the phage-host interactions provides insight into the phage-host interaction and

dynamics and may lead to new strategies for the development and application of successful

phage therapy [114, 115]. Furthermore, the understanding of the interactions makes it possible

adapt to phage selection towards the desired outcome [116]. This includes selecting optimal

phage(s) that can overcome host phage-resistance mechanisms, select for attenuated virulence,

for impaired fitness/growth, and/or select for increased susceptibility to antimicrobials. Further

studies comparing how different phages select for resistant bacteria may also lead to better

understanding on how bacteria react on phage infection. Although the full complexity of the

interactions cannot be captured, in vitro experiments can still provide essential information

needed for further application in a therapeutic setting (in vivo/in situ) [117].

For 44 phage-resistant strains no detected genomic changes differentiated them from the WT

strain. This could be caused by both laboratory issues, such as non-resistant strains were

erroneously defined as true resistant mutants based on AUC values, or actual variations that


166

were missed due to genetic variation within discarded repetitive regions identified by NUCmer

or partly loss off genes of which the consequence on the overall gene function were not

investigated. Our experiments were conducted in vitro and thus caution should be used when

interpreting our findings for in vivo applications. The co-evolutionary interactions, including

phage resistance, observed in laboratory experiments can differ from the highly complex

interactions found in natural environments, which may influence the ecology and evolution of

both phages and their hosts [118].

In conclusion, under selective pressure of virulent phages, bacterial strains of E. coli can

acquire one or more spontaneous mutations or gene losses that confer phage resistance in vitro.

The majority of detected phage-resistant mutant strains from this study were shown to resist

phage infection through mechanisms related to phage adsorption inhibition. Interestingly, we

also found several new genes, including two encoding hypothetical proteins, that could

potentially play a role in E. coli phage resistance. There were no indications that the infecting

phages were able to overcome the phage resistance. Nevertheless, as the initial infection

targeted known E. coli virulence factors, such as OMPs and the LPS, and thus, potentially

decreased the APEC virulence, the infecting phage still possessed desirable traits for phage

therapy application. Furthermore, in many cases phage resistance was associated with fitness

cost for the affected mutant strain resulting in up to ~65% decrease in growth. Thus, this study

provides valuable information about the interactions between virulent coliphages and their

host, which may aid prediction of the phage-host interaction outcome and future development

of a successful phage therapy.


167

References



virology-100114-054952.





[4] Buckling, A.; Rainey, P. B. Antagonistic Coevolution between a Bacterium and a Bact eriophage.










[8] Weissman, J. L.; Holmes, R.; Barrangou, R.; Moineau, S.; Fagan, W. F.; Levin, B.; Johnson, P. L.

F. Immune Loss as a Driver of Coexistence during Host-Phage Coevolution. ISME J., 2018, 12 (2),

585–597. https://doi.org/10.1038/ismej.2017.194.

[9] Koskella, B.; Brockhurst, M. A. Bacteria –Phage Coevolution as a Driver of Ecological and

Evolutionary Processes in Microbial Communities. FEMS Microbiol. Rev., 2014, 38 (5), 916–931.

https://doi.org/10.1111/1574-6976.12072.

[10] Azam, A. H.; Tanji, Y. Bacteriophage-Host Arm Race: An Update on the Mechanism of Phage

Resistance in Bacteria and Revenge of the Phage with the Perspective for Phage Therapy. Appl.

Microbiol. Biotechnol., 2019, 103, 2121–2131. https://doi.org/10.1007/s00253-019-09629-x.

[11] Burmeister, A. R.; Turner, P. E. Trading-off and Trading-up in the World of Bacteria –Phage

Evolution. Curr. Biol., 2020, 30 (19), R1120–R1124. https://doi.org/10.1016/j.cub.2020.07.036.

[12] Vasu, K.; Nagaraja, V. Diverse Functions of Restriction-Modification Systems in Addition to

Cellular Defense. Microbiol. Mol. Biol. Rev., 2013, 77 (1), 53–72.


[13] Vale, P. F.; Lafforgue, G.; Gatchitch, F.; Gardan, R.; Moineau, S.; Gandon, S. Costs of CRISPR-










[16] Dho-Moulin, M.; Fairbrother, J. M. Avian Pathogenic Escherichia Coli (APEC). Vet. Res., 1999, 30

(2–3), 299–316.



168



[18] Nolan, L. K.; Barnes, H. J.; Vaillancourt, J.; Abdul-aziz, T.; Logue, C. M. Colibacillosis. In Diseases

of Poultry; Swayne, D. E., Ed.; John Wiley & Sons, Inc., 2013; pp 751–805.

https://doi.org/10.1002/9781119421481.

[19] Kathayat, D.; Helmy, Y. A.; Debla is, L.; Rajashekara, G. Novel Small Molecules Affecting Cell

Membrane as Potential Therapeutics for Avian Pathogenic Escherichia Coli. Sci. Rep., 2018, 8 (1),

1–16. https://doi.org/10.1038/s41598-018-33587-5.

[20] Centers for Disease Control and Prevention (CDC). Antibiotic Resistance Threats in the United

States; 2013.

[21] World Health Organization (WHO). Antimicrobial Resistance: Global Report on Surveillance;

Geneva, Switzerland, 2014. https://doi.org/10.1007/s13312-014-0374-3.

[22] Lin, D. M.; Koskella , B.; Lin, H. C. Phage Therapy: An Alternative to Antibiotics in the Age of





[24] Colson, C.; Glover, S. W.; Symonds, N.; Stacey, K. A. The Location of the Genes for Host -

Controlled Modification and Restriction in Escherichia Coli K-12. Genetics, 1965, 52 (5), 1043–

1050.

[25] Sørensen, P. E.; Van Den Broeck, W.; Kiil, K.; Jasinskyte, D.; Moodley, A.; Garmyn, A.; Ingmer,

H.; Butaye, P. New Insights into the Biodiversity of Coliphages in the Intestine of Poultry. Sci. Rep.,

2020, 10, 15220. https://doi.org/10.1038/s41598-020-72177-2.

[26] Reinheimer, J. A.; Morelli, L.; Bottazzi, V.; Suárez, V. Phenotypic Variability among Cells of

Lactobacillus Helveticus ATCC 15807. Int. Dairy J., 1995, 5, 97–103.

[27] Carminati, D.; Zennaro, R.; Neviani, E.; Giraffa, G. Selezione e Caratteristiche Di Mutanti Fago-

Resistenti Di Lactobacillus Helveticus. Sci. e Tec. Latt., 1993, 44 (1), 33–48.

[28] Capra, M. L.; Mercanti, D. J.; Rossetti, L. C.; Reinheimer, J. A.; Quiberoni, A. Isolation and

Phenotypic Characterization of Lactobacillus Casei and Lactobacillus Paracasei Bacteriophage-

Resistant Mutants. J. Appl. Microbiol., 2011, 111 (2), 371–381. https://doi.org/10.1111/j.1365-

2672.2011.05056.x.

[29] Kropinski, A. M.; Mazzocco, A.; Waddell, T. E.; Lingohr, E.; Johnson, R. P. Enumeration of

Bacteriophages by Double Agar Overlay Plaque Assay. In Bacteriophages. Methods in Molecular

Biology; Clokie, M. R., Kropinski, A. M., Eds.; Humana Press, 2009; Vol. 501, pp 69 –76.

https://doi.org/10.1007/978-1-60327-164-6_7.



https://doi.org/10.3390/v10040189.




[32] Bankevich, A.; Nurk, S.; Antipov, D.; Gurevich, A. A.; Dvorkin, M.; Kulikov, A. S.; Lesin, V. M.;

Nikolenko, S. I.; Pham, S.; Prjibelski, A. D.; et al. SPAdes: A New Genome Assembly Algorithm

and Its Applications to Single-Cell Sequencing. J. Comput. Biol., 2012, 19 (5), 455–477.

https://doi.org/10.1089/cmb.2012.0021.




169


[34] Seemann, T. Prokka: Rapid Prokaryotic Genome Annotation. Bioinformatics, 2014, 30 (14), 2068–

2069. https://doi.org/10.1093/bioinformatics/btu153.

[35] Page, A. J.; Cummins, C. A.; Hunt, M.; Wong, V. K.; Reuter, S.; Holden, M. T. G.; Fookes, M.;

Falush, D.; Keane, J. A.; Parkhill, J. Roary: Rapid Large-Scale Prokaryote Pan Genome Analysis.

Bioinformatics, 2015, 31 (22), 3691–3693. https://doi.org/10.1093/bioinformatics/btv421.

[36] Bradley, P.; Gordon, N. C.; Walker, T. M.; Dunn, L.; Heys, S.; Huang, B.; Earle, S.; Pankhurst, L.

J.; Anson, L.; De Cesare, M.; et al. Rapid Antibiotic-Resistance Predictions from Genome Sequence

Data for Staphylococcus Aureus and Mycobacterium Tuberculosis. Nat. Commun., 2015, 6, 1–14.

https://doi.org/10.1038/ncomms10063.




[38] Feldgarden, M.; Brover, V.; Haft, D. H.; Prasad, A. B.; Slotta, D. J.; Tolstoy, I.; Tyson, G. H.; Zhao,

S.; Hsu, C. H.; McDermott, P. F.; et al. Validating the AMRFINder Tool and Resistan ce Gene

Database by Using Antimicrobial Resistance Genotype-Phenotype Correlations in a Collection of

Isolates. Antimicrob. Agents Chemother., 2019, 63 (11), 1–19. https://doi.org/10.1128/AAC.00483-

19.

[39] Alcock, B. P.; Raphenya, A. R.; Lau, T. T. Y.; Tsang, K. K.; Bouchard, M.; Edalatmand, A.; Huynh,

W.; Nguyen, A.-L. V; Cheng, A. A.; Liu, S.; et al. CARD 2020: Antibiotic Resistome Surveillance

with the Comprehensive Antibiotic Resistance Database. Nucleic Acids Res., 2019, 48 (D1), D517–

D525. https://doi.org/10.1093/nar/gkz935.

[40] Couvin, D.; Bernheim, A.; Toffano-Nioche, C.; Touchon, M.; Michalik, J.; Néron, B.; Rocha, E. P.

C.; Vergnaud, G.; Gautheret, D.; Pourcel, C. CRISPRCasFinder, an Update of CRISRFinder,

Includes a Portable Version, Enhanced Performance and Integrates Search for Cas Proteins. Nucleic

Acids Res., 2018, 46 (W1), W246–W251. https://doi.org/10.1093/nar/gky425.

[41] Sahl, J. W.; Lemmer, D.; Travis, J.; Schupp, J. M.; Gillece, J. D.; Aziz, M.; Driebe, E. M.; Drees, K.

P.; Hicks, N. D.; Williamson, C. H. D.; et al. NASP: An Accurate, Rapid Method for the

Identification of SNPs in WGS Datasets That Supports Flexible Input and Output Formats. Microb.

genomics, 2016, 2 (8), e000074. https://doi.org/10.1099/mgen.0.000074.

[42] Li, H.; Durbin, R. Fast and Accurate Short Read Alignment with Burrows-Wheeler Transform.

Bioinformatics, 2009, 25 (14), 1754–1760. https://doi.org/10.1093/bioinformatics/btp324.

[43] Marçais, G.; Delcher, A. L.; Phillippy, A. M.; Coston, R.; Salzberg, S. L.; Zimin, A. MUMmer4: A

Fast and Versatile Genome Alignment System. PLoS Comput. Biol., 2018, 14 (1), e1005944.

https://doi.org/10.1371/journal.pcbi.1005944.

[44] Steenbergen, S. M.; Vimr, E. R. Biosynthesis of the Escherichia Coli K1 Group 2 Polysialic Acid

Capsule Occurs within a Protected Cytoplasmic Compartment. Mol. Microbiol., 2008, 68 (5), 1252–

1267. https://doi.org/10.1111/j.1365-2958.2008.06231.x.

[45] Cerveny, L.; Straskova, A.; Dankova, V.; Hartlova, A.; Ceckova, M.; Staud, F.; Stulik, J.

Tetratricopeptide Repeat Motifs in the World of Bacterial Pathogens: Role in Virulence Mechanisms.

Infect. Immun., 2013, 81 (3), 629–635. https://doi.org/10.1128/IAI.01035-12.

[46] Schütze, A.; Benndorf, D.; Püttker, S.; Kohrs, F.; Bettenbrock, K. The Impact of AckA, Pta, and

AckA-Pta Mutations on Growth, Gene Expression and Protein Acetylation in Escherichia Coli K-

12. Front. Microbiol., 2020, 11 (February), 1–13. https://doi.org/10.3389/fmicb.2020.00233.

[47] Deochand, D. K.; Grove, A. MarR Family Transcription Factors: Dynamic Variations on a Common

Scaffold. Crit. Rev. Biochem. Mol. Biol., 2017, 52 (6), 595–613.

https://doi.org/10.1080/10409238.2017.1344612.


170

[48] Cowley, L. A.; Low, A. S.; Pickard, D.; Boinett, C. J.; Dallman, T. J.; Day, M.; Perry, N.; Gally, D.

L.; Parkhill, J.; Jenkins, C.; et al. Transposon Insertion Sequencing Elucidates Novel Gene

Involvement in Susceptibility and Resistance to Phages T4 and T7 in Escherichia Coli O157. MBio,

2018, 9 (4), e00705-18. https://doi.org/10.1128/mBio.00705-18.

[49] Nagel, G. M.; Cumberledge, S.; Johnson, M. S.; Petrella, E.; Weber, B. H. The β Subunit of E. Coil

Glycyl-TRNA Synthetase Plays a Major Role in TRNA Recognition. Nucleic Acids Res., 1984, 12

(10), 4377–4384. https://doi.org/10.1093/nar/12.10.4377.

[50] Campbell, J. W.; Cronan, J. E. The Enigmatic Escherichia Coli FadE Gene Is YafH. J. Bacteriol.,

2002, 184 (13), 3759–3764. https://doi.org/10.1128/JB.184.13.3759-3764.2002.

[51] Peracchi, A.; Veiga-Da-Cunha, M.; Kuhara, T.; Ellens, K. W.; Paczia, N.; Stroobant, V.; Seliga, A.

K.; Marlaire, S.; Jaisson, S.; Bommer, G. T.; et al. Nit1 Is a Metabolite Repair Enzyme That

Hydrolyzes Deaminated Glutathione. Proc. Natl. Acad. Sci. U. S. A., 2017, 114 (16), E3233–E3242.

https://doi.org/10.1073/pnas.1613736114.

[52] Deckers, D.; Vanlint, D.; Callewaert, L.; Aertsen, A.; Michiels, C. W. Role of the Lysozyme Inhibitor

Ivy in Growth or Survival of Escherichia Coli and Pseudomonas Aeruginosa Ba cteria in Hen Egg

White and in Human Saliva and Breast Milk. Appl. Environ. Microbiol., 2008, 74 (14), 4434–4439.

https://doi.org/10.1128/AEM.00589-08.

[53] Zhang, Y.; Pohlmann, E. L.; Serate, J.; Conrad, M. C.; Roberts, G. P. Mutagenesis and Functional

Characterization of the Four Domains of GlnD, a Bifunctional Nitrogen Sensor Protein. J. Bacteriol.,

2010, 192 (11), 2711–2721. https://doi.org/10.1128/JB.01674-09.

[54] Daines, D. A.; Wright, L. F.; Chaffin, D. O.; Rubens, C. E.; Silver, R. P. NeuD Plays a Role in the

Synthesis of Sialic Acid in Escherichia Coli K1. FEMS Microbiol. Lett., 2000, 189 (2), 281–284.

https://doi.org/10.1111/j.1574-6968.2000.tb09244.x.

[55] Wright, R. C. T.; Friman, V. P.; Smith, M. C. M.; Brockhurst, M. A. Resistance Evolution against

Phage Combinations Depends on the Timing and Order of Exposure. MBio, 2019, 10 (5), e01652-


[56] Bertozzi Silva, J.; Storms, Z.; Sauvageau, D. Host Receptors for Bacteriophage Adsorption. FEMS


[57] Smith, S. G. J.; Mahon, V.; Lambert, M. A.; Fagan, R. P. A Molecular Swiss Army Knife: OmpA

Structure, Function and Expression. FEMS Microbiol. Lett., 2007, 273 (1), 1–11.

https://doi.org/10.1111/j.1574-6968.2007.00778.x.

[58] Korch, S. B.; Henderson, T. A.; Hill, T. M. Characterization of the HipA7 Allele of Escherichia Coli

and Evidence That High Persistence Is Governed by (p)PpGpp Synthesis. Mol. Microbiol., 2003, 50

(4), 1199–1213. https://doi.org/10.1046/j.1365-2958.2003.03779.x.

[59] Valentini, G.; Chiarelli, L.; Fortin, R.; Speranza, M. L.; Galizzi, A.; Mattevi, A. The Allosteric

Regulation of Pyruvate Kinase. J. Biol. Chem., 2000, 275 (24), 18145–18152.

https://doi.org/10.1074/jbc.M001870200.

[60] Wang, Z.; Wang, J.; Ren, G.; Li, Y.; Wang, X. Influence of Core Oligosaccharide of

Lipopolysaccharide to Outer Membrane Behavior of Escherichia Coli. Mar. Drugs, 2015, 13 (6),

3325–3339. https://doi.org/10.3390/md13063325.

[61] Pedersen, K.; Gerdes, K. Multiple Hok Genes on the Chromosome of Escherichia Coli. Mol.

Microbiol., 1999, 32 (5), 1090–1102. https://doi.org/10.1046/j.1365-2958.1999.01431.x.

[62] Frederix, M.; Hütter, K.; Leu, J.; Batth, T. S.; Turner, W. J.; Rüegg, T. L.; Blanch, H. W.; Simmons,

B. A.; Adams, P. D.; Keasling, J. D.; et al. Development of a Native Escherichia Coli Induction

Dystem for Ionic Liquid Tolerance. PLoS One, 2014, 9 (7), e101115.


[63] Stach, N.; Kalinska, M.; Zdzalik, M.; Kitel, R.; Karim, A.; Serwin, K.; Rut, W.; Larsen, K.; Jabaiah,


171

A.; Firlej, M.; et al. Unique Substrate Specificity of SplE Serine Protease from Staphylococcus

Aureus. Structure, 2018, 26 (4), 572-579.e4. https://doi.org/10.1016/j.str.2018.02.008.

[64] Grauschopf, U.; Fritz, A.; Glockshuber, R. Mechanism of the Electron Transfer Catalyst DsbB from

Escherichia Coli. EMBO J., 2003, 22 (14), 3503–3513. https://doi.org/10.1093/emboj/cdg356.

[65] Stenberg, F.; Chovanec, P.; Maslen, S. L.; Robinson, C. V.; Ilag, L. L.; von Heijne, G.; Daley, D. O.

Protein Complexes of the Escherichia Coli Cell Envelope. J. Biol. Chem., 2005, 280 (41), 34409–

34419. https://doi.org/10.1074/jbc.M506479200.

[66] Deniziak, M. A.; Barciszewski, J. Methionyl-TRNA Synthetase. Acta Biochim. Pol., 2001, 48 (2),

337–350. https://doi.org/10.18388/abp.2001_3919.

[67] Shepherd, J.; Ibba, M. Bacterial Transfer RNAs. FEMS Microbiol. Rev., 2015, 39 (3), 280–300.

https://doi.org/10.1093/femsre/fuv004.

[68] Noriega, C. E.; Lin, H.-Y.; Chen, L.-L.; Williams, S. B.; Stewart, V. Asymmetric Cross-Regulation

between the Nitrate-Responsive NarX-NarL and NarQ-NarP Two-Component Regulatory Systems

from Escherichia Coli K-12. Mol. Microbiol., 2010, 75 (2), 394–412. https://doi.org/10.1111/j.1365-

2958.2009.06987.x.

[69] Breton, R.; Sanfacon, H.; Papayannopoulos, I.; Biemann, K.; Lapointe, J. Glutamyl-TRNA

Synthetase of Escherichia Coli. Isolation and Primary Structure of the GltX Gene and Homology

with Other Aminoacyl-TRNA Synthetases. J. Biol. Chem., 1986, 261 (23), 10610–10617.

https://doi.org/10.1016/s0021-9258(18)67429-0.

[70] Navarro-Garcia, F.; Ruiz-Perez, F.; Cataldi, Á.; Larzábal, M. Type VI Secretion System in

Pathogenic Escherichia Coli: Structure, Role in Virulence, and Acquisition. Front. Microbiol., 2019,

10, 1965. https://doi.org/10.3389/fmicb.2019.01965.

[71] Zhou, Y.; Bao, H.; Zhang, H.; Pang, M.; Zhu, S.; Wang, R. A Spontaneous RapZ Mutant Impairs

Infectivity of Lytic Bacteriophage VB_EcoM_JS09 against En terotoxigenic Escherichia Coli.

mSphere, 2021, 6, e01286-20. https://doi.org/10.1128/mSphere.01286-20.

[72] Mutalik, V. K.; Adler, B. A.; Rishi, H. S.; Piya, D.; Zhong, C.; Koskella, B.; Kutter, E. M.; Calendar,

R.; Novichkov, P. S.; Price, M. N.; et al. High-Throughput Mapping of the Phage Resistance

Landscape in E. Coli. PLOS Biol., 2020, 18 (10), e3000877.


[73] Khan, M. A.; Durica‐Mitic, S.; Göpel, Y.; Heermann, R.; Görke, B. Small RNA‐binding Protein

RapZ Mediates Cell Envelope Precursor Sensing and Signaling in Escherichia Coli. EMBO J., 2020,

39, e103848. https://doi.org/10.15252/embj.2019103848.

[74] Keseler, I. M.; Mackie, A.; Santos-Zavaleta, A.; Billington, R.; Bonavides-Martínez, C.; Caspi, R.;

Fulcher, C.; Gama-Castro, S.; Kothari, A.; Krummenacker, M.; et al. The EcoCyc Database:

Reflecting New Knowledge about Escherichia Coli K-12. Nucleic Acids Res., 2017, 45 (D1), D543–

D550. https://doi.org/10.1093/nar/gkw1003.

[75] Di Luccio, E.; Petschacher, B.; Voegtli, J.; Chou, H.; Stahlberg, H.; Nidetzky, B.; Wilson, D. K.

Structural and Kinetic Studies of Induced Fit in Xylulose Kinase from Escherichia Coli. J. Mol. Biol.,

2007, 365 (3), 783–798. https://doi.org/10.1016/j.jmb.2006.10.068.

[76] Seo, S. W.; Kim, D.; Latif, H.; O’Brien, E. J.; Szubin, R.; Palsson, B. O. Deciphering Fur

Transcriptional Regulatory Network Highlights Its Complex Role beyond Iron Metabolism in

Escherichia Coli. Nat. Commun., 2014, 5, 4910. https://doi.org/10.1038/ncomms5910.

[77] Terakawa, A.; Natsume, A.; Okada, A.; Nishihata, S.; Kuse, J.; Tanaka, K.; Takenaka, S.; Ishikawa,

S.; Yoshida, K. Bacillus Subtilis 5′-Nucleotidases with Various Functions and Substrate

Specificities. BMC Microbiol., 2016, 16, 249. https://doi.org/10.1186/s12866-016-0866-5.

[78] Alicea, I.; Marvin, J. S.; Miklos, A. E.; Ellington, A. D.; Looger, L. L.; Schreiter, E. R. Structure of

the Escherichia Coli Phosphonate Binding Protein PhnD and Rationally Optimized Phosphonate


172

Biosensors. J. Mol. Biol., 2011, 414 (3), 356–369. https://doi.org/10.1016/j.jmb.2011.09.047.

[79] Al Mamun, A. A. M.; Lombardo, M.-J.; Shee, C.; Lisewski, A. M.; Gonzalez, C.; Lin, D.; Nehring,

R. B.; Saint-Ruf, C.; Gibson, J. L.; Frisch, R. L.; et al. Identity and Function of a Large Gene Network

Underlying Mutagenic Repair of DNA Breaks. Science (80-. )., 2012, 338 (6112), 1344–1348.

https://doi.org/10.1126/science.1226683.

[80] Khemici, V.; Poljak, L.; Luisi, B. F.; Carpousis, A. J. The RNase E of Escherichia Coli Is a

Membrane-Binding Protein. Mol. Microbiol., 2008, 70 (4), 799–813. https://doi.org/10.1111/j.1365-

2958.2008.06454.x.

[81] Magnet, S.; Courvalin, P.; Lambert, T. Activation of the Cryptic Aac(6’)-Iy Aminoglycoside

Resistance Gene of Salmonella by a Chromosomal Deletion Generating a Transcriptional Fusion. J.


[82] Sunthornthummas, S.; Doi, K.; Rangsiruji, A.; Krajangsung, S.; Sarawaneeyaruk, S.; Pringsulaka,

O. Isolation and Cha racterization of Spontaneous Phage-Resistant Mutants of Lactobacillus

Paracasei. Food Control, 2019, 99 (September 2018), 114–123.

https://doi.org/10.1016/j.foodcont.2018.12.037.

[83] Guglielmotti, D. M.; Reinheimer, J. A.; Binetti, A. G.; Giraffa, G.; Carminati, D.; Quiberoni, A.

Characterization of Spontaneous Phage-Resistant Derivatives of Lactobacillus Delbrueckii

Commercial Strains. Int. J. Food Microbiol., 2006, 111 (2), 126–133.

https://doi.org/10.1016/j.ijfoodmicro.2006.04.035.

[84] Binetti, A. G.; Bailo, N. B.; Reinheimer, J. A. Spontaneous Phage-Resistant Mutants of

Streptococcus Thermophilus: Isolation and Technological Characteristics. Int. Dairy J., 2007, 17 (4),

343–349. https://doi.org/10.1016/j.idairyj.2006.05.002.

[85] Rostøl, J. T.; Marraffini, L. (Ph)Ighting Phages: How Bacteria Resist Their Parasites. Cell Host

Microbe, 2019, 25 (2), 184–194. https://doi.org/10.1016/j.chom.2019.01.009.

[86] Hantke, K. Compilation of Escherichia Coli K-12 Outer Membrane Phage Receptors – Their

Function and Some Historical Remarks. FEMS Microbiol. Lett., 2020, 367 (2), fnaa013.

https://doi.org/10.1093/femsle/fnaa013.

[87] Wang, Y. The Function of OmpA in Escherichia Coli. Biochem. Biophys. Res. Commun., 2002, 292

(2), 396–401. https://doi.org/10.1006/bbrc.2002.6657.



https://doi.org/10.1128/IAI.00926-19.





[90] Salazar, K. C.; Ma, L.; Green, S. I.; Zulk, J. J.; Trautner, B. W.; Ramig, R. F.; Clark, J. R.;

Terwilliger, A. L.; Maresso, A. W. Antiviral Resistance and Phage Counter Adaptation to Antib iotic-

Resistant Extraintestinal Pathogenic Escherichia Coli. MBio, 2021, 12 (2), e00211-21.


[91] Bulieris, P. V.; Behrens, S.; Holst, O.; Kleinschmidt, J. H. Folding and Insertion of the Outer

Membrane Protein OmpA Is Assisted by the Chaperone Skp and by Lipopolysaccharide. J. Biol.

Chem., 2003, 278 (11), 9092–9099. https://doi.org/10.1074/jbc.M211177200.

[92] Islam, M. Z.; Fokine, A.; Mahalingam, M.; Zhang, Z.; Garcia -Doval, C.; Van Raaij, M. J.; Rossmann,

M. G.; Rao, V. B. Molecular Anatomy of the Receptor Binding Module of a Bacteriophage Long

Tail Fiber. PLoS Pathog., 2019, 15 (12), e1008193. https://doi.org/10.1371/journal.ppat.1008193.

[93] Chen, P.; Sun, H.; Ren, H.; Liu, W.; Li, G.; Zhang, C. LamB, OmpC, and the Core


173

Lipopolysaccharide of Escherichia Coli K-12 Function as Receptors of Bacteriophage Bp7. J. Virol.,

2020, 94 (12), e00325-20. https://doi.org/10.1128/JVI.00325-20.

[94] Pagnout, C.; Sohm, B.; Razafitianamaharavo, A.; Caillet, C.; Offroy, M.; Leduc, M.; Gendre, H.;

Jomini, S.; Beaussart, A.; Bauda, P.; et al. Pleiotropic Effects of Rfa -Gene Mutations on Escherichia

Coli Envelope Properties. Sci. Rep., 2019, 9 (1), 9696. https://doi.org/10.1038/s41598-019-46100-3.

[95] Borin, J. M.; Avrani, S.; Barrick, J. E.; Petrie, K. L.; Meyer, J. R. Coevolutionary Phage Training

Leads to Greater Bacterial Suppression and Delays the Evolution of Phage Resistance. Proc. Natl.

Acad. Sci. U. S. A., 2021, 118 (23), e2104592118. https://doi.org/10.1073/pnas.2104592118.

[96] Gonzalez, G. M.; Durica -Mitic, S.; Hardwick, S. W.; Moncrieffe, M. C.; Resch, M.; Neumann, P.;

Ficner, R.; Görke, B.; Luisi, B. F. Structural Insights into RapZ-Mediated Regulation of Bacterial

Amino-Sugar Metabolism. Nucleic Acids Res., 2017, 45 (18), 10845–10860.

https://doi.org/10.1093/nar/gkx732.

[97] Azam, A. H.; Hoshiga, F.; Takeuchi, I.; Miyanaga, K.; Tanji, Y. Analysis of Phage Resistance in

Staphylococcus Aureus SA003 Revea ls Different Binding Mechanisms for the Closely Related

Twort-like Phages ɸSA012 and ɸSA039. Appl. Microbiol. Biotechnol., 2018, 102 (20), 8963–8977.

https://doi.org/10.1007/s00253-018-9269-x.

[98] Vimr, E. R.; Kalivoda, K. A.; Deszo, E. L.; Steenbergen, S. M. Diversity of Microbial Sialic Acid

Metabolism. Microbiol. Mol. Biol. Rev., 2004, 68 (1), 132–153.

https://doi.org/10.1128/MMBR.68.1.132-153.2004.

[99] Silver, R. P.; Prior, K.; Nsahlai, C.; Wright, L. F. ABC Transporters and the Export of Capsular

Polysaccharides from Gram-Negative Bacteria. Res. Microbiol., 2001, 152 (3–4), 357–364.

https://doi.org/10.1016/S0923-2508(01)01207-4.

[100] Schwarzer, D.; Buettner, F. F. R.; Browning, C.; Nazarov, S.; Rabsch, W.; Bethe, A.; Oberbeck, A.;

Bowman, V. D.; Stummeyer, K.; Muhlenhoff, M.; et al. A Multivalent Adsorption Apparatus

Explains the Broad Host Range of Phage Phi92: A Comprehensive Genomic and Structural Analysis.

J. Virol., 2012, 86 (19), 10384–10398. https://doi.org/10.1128/jvi.00801-12.

[101] Latka, A.; Maciejewska, B.; Majkowska-Skrobek, G.; Briers, Y.; Drulis-Kawa, Z. Bacteriophage-

Encoded Virion-Associated Enzymes to Overcome the Carbohydrate Barriers during the Infection

Process. Appl. Microbiol. Biotechnol., 2017, 101 (8), 3103–3119. https://doi.org/10.1007/s00253-

017-8224-6.

[102] Scholl, D.; Adhya, S.; Merril, C. Escherichia Coli K1’s Capsule Is a Barrier to Bacteriophage T7.

Appl. Environ. Microbiol., 2005, 71 (8), 4872–4874. https://doi.org/10.1128/AEM.71.8.4872-

4874.2005.

[103] Wang, Z.; Zhao, C.; Wang, Y.; Sun, Z.; Wang, N. PANDA: Protein Function Prediction Using

Domain Architecture and Affinity Propagation. Sci. Rep., 2018, 8 (1), 3484.

https://doi.org/10.1038/s41598-018-21849-1.

[104] Goldberg, T.; Hecht, M.; Hamp, T.; Karl, T.; Yachdav, G.; Ahmed, N.; Altermann, U.; Angerer, P.;

Ansorge, S.; Balasz, K.; et al. LocTree3 Prediction of Localization. Nucleic Acids Res., 2014, 42

(W1), W350–W355. https://doi.org/10.1093/nar/gku396.

[105] Hu, P.; Janga, S. C.; Babu, M.; Díaz-Mejía, J. J.; Butland, G.; Yang, W.; Pogoutse, O.; Guo, X.;

Phanse, S.; Wong, P.; et al. Global Functional Atlas of Escherichia Coli Encompassing Previously

Uncharacterized Proteins. PLoS Biol., 2009, 7 (4), e1000096.


[106] Plotkin, J. B.; Kudla, G. Synonymous but Not the Same: The Causes and Consequences of Codon

Bias. Nat. Rev. Genet., 2011, 12 (1), 32–42. https://doi.org/10.1038/nrg2899.

[107] Pourcel, C.; Touchon, M.; Villeriot, N.; Vernadet, J.-P.; Couvin, D.; Toffano-Nioche, C.; Vergnaud,

G. CRISPRCasdb a Successor of CRISPRdb Containing CRISPR Arrays and Cas Genes from

Complete Genome Sequences, and Tools to Download and Query Lists of Repeats and Spacers.


174

Nucleic Acids Res., 2020, 48 (D1), D535–D544. https://doi.org/10.1093/nar/gkz915.

[108] Denes, T.; Den Bakker, H. C.; Tokman, J. I.; Guldimann, C.; Wiedmann, M. Selection and

Characterization of Phage-Resistant Mutant Strains of Listeria Monocytogenes Reveal Host Genes

Linked to Phage Adsorption. Appl. Environ. Microbiol., 2015, 81 (13), 4295–4305.

https://doi.org/10.1128/AEM.00087-15.

[109] Levin, B. R.; Moineau, S.; Bushman, M.; Barrangou, R. The Population and Evolutionary Dynamics

of Phage and Bacteria with CRISPR–Mediated Immunity. PLoS Genet., 2013, 9 (3), e1003312.




[111] Meyer, J. R.; Dobias, D. T.; Weitz, J. S.; Barrick, J. E.; Quick, R. T.; Lenski, R. E. Repeatability and

Contingency in the Evolution of a Key Innovation in Phage Lambda. Science (80-. )., 2012, 335

(6067), 428–432. https://doi.org/10.1126/science.1214449.

[112] Wandro, S.; Oliver, A.; Gallagher, T.; Weihe, C.; England, W.; Martiny, J. B. H.; Whiteson, K.

Predictable Molecular Adaptation of Coevolving Enterococcus Faecium and Lytic Phage EfV12 -

Phi1. Front. Microbiol., 2019, 9, 3192. https://doi.org/10.3389/fmicb.2018.03192.

[113] Hall, A. R.; Scanlan, P. D.; Buckling, A. Bacteria‐phage Coevolution and the Emergence of

Generalist Pathogens. Am. Nat., 2011, 177 (1), 44–53. https://doi.org/10.1086/657441.

[114] Chaturongakul, S.; Ounjai, P. Phage-Host Interplay: Examples from Tailed Phages and Gram -

Negative Bacterial Pa thogens. Front. Microbiol., 2014, 5, 442.


[115] Federici, S.; Nobs, S. P.; Elinav, E. Phages and Their Potential to Modulate the Microbiome and

Immunity. Cell. Mol. Immunol., 2021, 18 (4), 889–904. https://doi.org/10.1038/s41423-020-00532-

4.







https://doi.org/10.1038/s41467-017-00158-7.


175

Su

pp

lem

en

tary F

igu

re

S1

. E

xperi

menta

l se

tup o

f th

e is

ola

tio

n o

f p

hag

e r

esi

stant

AP

EC

str

ain

s b

y s

eco

ndary

cu

lture

(S

C)

or

agar

pla

te (

AP

) m

eth

ods.

At

tim

e 0

h,

susc

ep

tib

le A

M621

WT

cu

lture

s w

ere

infe

cte

d w

ith

phage a

t fo

ur d

iffe

rent m

ult

iplicit

ies

of

infe

ctio

n (M

OIs

): M

OI

0.1

, MO

I 1

, M

OI

10

, and M

OI

100

. An

un

-infe

cted

co

ntr

ol (A

M6

21)

and n

ega

tiv

e co

ntr

ols

(P

hage a

nd L

B)

were

in

clu

ded. A

P a

pp

roach

: su

spen

sion

s w

ere

str

eaked d

irect

ly o

nto

LB

aga

r p

late

s a

nd in

cubate

d f

or

48 h

.

Resi

sta

nt

co

lon

ies

were

pic

ked

aft

er 2

4 h

and

48 h

of

incu

batio

n, gro

wn

in

LB

bro

th a

nd p

uri

fied b

y t

hre

e c

onse

cutiv

e s

trea

kin

gs

on

LB

agar

pla

tes.

SC

appro

ach

:

Rem

ain

ing “

0h” s

usp

ensi

ons

were

incu

bate

d f

or 5

h. S

usp

en

sion

s ex

hib

itin

g ly

sis

were

in

cu

bate

d f

or a

ddit

ional 2

4 h

. If

sub

seq

uent (s

econdary

) gro

wth

wa

s o

bse

rved

,

susp

en

sio

ns

were

str

eak

ed o

nto

pla

tes

an

d p

roces

sed

as

desc

rib

ed f

or th

e A

P a

pp

roa

ch. P

hage resi

stant st

rain

s w

ere

nam

ed a

cco

rdin

g to

ex

peri

menta

l setu

p (

AP

or SC

an

d M

OI.

Ex

per

imen

tal s

teps

follo

win

g th

e is

ola

tion a

re o

utlin

ed in

gre

en

.


176

Supplementary Figure S2. Phage-resistant strains isolated from pathogenic E. coli by secondary culture (SC) or

agar plate (AP) methods. (A) Number of resistant strains for each method and multiplicity of infection (MOI)

using the AP or SC method, including MOI 0.1 (black), MOI 1 (dark grey), MOI 10 (light grey), and MOI 100

(white). (B) Number of resistant strains isolated for each timepoint (24 hours or 48 hours of incubation) for the

AP method (black) and SC method (grey). The number of resistant strains is indicated for each combination. The

numbers were obtained based on six repeated isolation experiments.

(A)

(B)


177

Su

pp

lem

en

tary

Ta

ble

S1

| A

rea u

nd

er

the c

urv

e (

AU

C)

an

d A

UC

decre

ase f

or

ph

age

-res

ista

nt st

rain

s re

lati

ve

to th

e W

T s

train

Mu

tan

t A

UC

D

ecrease

D

ecrea

se %

M

uta

nt

AU

C

Decrease

D

ecrea

se %

M

uta

nt

AU

C

Decrease

D

ecrea

se %

SC

48

_10

_8

5,7

1

0,6

6

5,2

S

C4

8_

01

_9

14

,7

1,6

9

,9

SC

24

_01

_11

15

,2

0,7

4

,4

AP

48

_1

_24

6

,0

8,7

5

9,1

A

P4

8_

10_2

4

13

,2

1,4

9

,8

SC

24

_1_

7

15

,6

0,7

4

,4

SC

24

_01

_5

6,8

9

,0

57

,0

SC

48

_1_

22

13

,2

1,4

9

,8

AP

48

_1

0_2

2

14

,1

0,5

3

,7

SC

48

_1_

9

10

,1

6,1

3

7,6

S

C2

4_

10

0_9

14

,8

1,5

9

,3

SC

24

_10

0_15

1

3,2

0

,5

3,7

S

C4

8_

1_

12

10

,4

5,5

3

4,6

A

P2

4_

100_

4

14

,4

1,4

8

,7

SC

24

_10

_14

14

,4

0,5

3

,6

AP

24

_1

0_1

4

9,2

4

,5

33

,0

SC

48

_10

0_7

14

,9

1,4

8

,5

SC

24

_10

_13

14

,4

0,5

3

,5

SC

24

_1_

9

11

,1

5,2

3

1,8

S

C2

4_

01

_15

13

,7

1,3

8

,4

AP

24

_1

0_2

4

14

,1

0,5

3

,5

SC

24

_1_

8

11

,1

5,1

3

1,6

S

C2

4_

1_

24

13

,5

1,2

8

,2

AP

48

_1

0_1

4

14

,5

0,5

3

,4

SC

48

_10

_23

10

,4

4,2

2

8,7

S

C2

4_

01

_13

13

,8

1,2

8

,1

AP

24

_1

00_

16

14

,5

0,5

3

,3

SC

24

_10

0_6

11

,5

4,3

2

7,1

S

C2

4_

10

_15

13

,8

1,2

7

,9

SC

48

_1_

10

15

,4

0,5

3

,2

SC

48

_10

_22

11

,1

3,5

2

4,0

S

C4

8_

10

_17

13

,8

1,2

7

,8

SC

24

_1_

16

14

,5

0,5

3

,2

AP

24

_1

00_

5

12

,3

3,5

2

2,0

S

C2

4_

01

_18

13

,8

1,2

7

,7

SC

24

_10

_16

14

,5

0,5

3

,0

SC

48

_10

_11

12

,9

3,0

1

8,8

S

C2

4_

1_

23

13

,6

1,1

7

,4

AP

48

_1

00_

11

15

,4

0,5

3

,0

SC

48

_10

_10

13

,0

2,9

1

8,1

S

C4

8_

1_

15

13

,9

1,1

7

,4

SC

24

_10

0_13

1

4,5

0

,4

2,9

AP

24

_1

0_5

1

3,1

2

,7

17

,1

SC

48

_10

_15

13

,9

1,1

7

,4

SC

48

_1_

16

14

,6

0,4

2

,3

SC

48

_1_

11

13

,2

2,6

1

6,7

S

C4

8_

10

_13

13

,9

1,1

7

,3

SC

48

_10

0_16

1

4,7

0

,3

1,9

SC

24

_10

_24

12

,2

2,4

1

6,6

S

C4

8_

01

_16

13

,9

1,1

7

,3

SC

24

_10

_17

14

,7

0,3

1

,9

SC

24

_01

_16

12

,5

2,4

1

6,2

S

C2

4_

01

_17

13

,9

1,1

7

,2

AP

24

_1

0_2

2

14

,4

0,3

1

,8

SC

24

_10

_10

13

,3

2,5

1

5,9

S

C4

8_

10

0_8

15

,1

1,2

7

,1

AP

24

_1

00_

23

14

,4

0,2

1

,6

SC

48

_01

_12

13

,3

2,5

1

5,9

A

P2

4_

10_2

3

13

,6

1,0

7

,0

SC

48

_10

_14

13

,5

0,1

1

,1

SC

24

_10

_23

12

,4

2,2

1

5,1

A

P2

4_

10_6

1

4,7

1

,1

6,8

A

P4

8_

100_

24

14

,5

0,1

1

,0

SC

48

_10

_24

12

,5

2,2

1

4,9

S

C2

4_

1_

13

14

,0

1,0

6

,8

AP

24

_1

_14

1

4,8

0

,2

1,0

AP

24

_100_18

12

,8

2,2

1

4,7

A

P2

4_

100_

8

15

,2

1,1

6

,5

SC

48

_10

0_22

1

4,5

0

,1

0,8

S

C2

4_

01

_14

11

,8

1,9

1

4,0

S

C4

8_

1_

13

14

,0

0,9

6

,2

AP

24

_1

00_

15

13

,6

0,1

0

,8

SC

24

_10

0_7

14

,0

2,2

1

3,8

A

P4

8_100_16

14

,0

0,9

6

,2

AP

24

_1

0_1

7

14

,9

0,1

0

,7

SC

24

_10

_22

12

,6

2,0

1

3,8

S

C2

4_100_18

14

,0

0,9

6

,1

AP

24

_1

0_1

3

13

,6

0,1

0

,6

SC

48

_01

_13

11

,8

1,8

1

3,4

S

C2

4_100_24

13

,8

0,9

5

,9

AP

48

_1

00_

14

13

,6

0,1

0

,6

AP

24

_100_12

13

,8

2,1

1

3,3

S

C2

4_

1_

14

14

,1

0,9

5

,7

SC

48

_01

_15

13

,6

0,1

0

,4

SC

48

_100_12

13

,9

2,0

1

2,4

A

P4

8_100_22

13

,8

0,8

5

,7

AP

48

_1

_13

1

4,9

0

,1

0,4

S

C2

4_100_12

13

,9

2,0

1

2,3

S

C4

8_

01

_14

14

,1

0,8

5

,6

SC

48

_10

0_24

1

4,6

0

,0

0,1

SC

24

_100_10

13

,9

1,9

1

2,3

S

C4

8_

01

_17

14

,1

0,8

5

,5

SC

48

_1_

14

13

,7

0,0

0

,1

SC

24

_1_

17

13

,2

1,8

1

1,9

S

C2

4_

1_

15

14

,1

0,8

5

,5

SC

48

_01

_23

14

,7

0,0

-0

,1

SC

24

_100_11

14

,0

1,8

1

1,6

S

C4

8_

1_

17

14

,2

0,8

5

,2

AP

48

_1

_22

1

4,7

-0

,1

-0,4

S

C2

4_

1_

22

13

,0

1,7

1

1,4

A

P4

8_

1_14

1

3,0

0

,7

4,9

S

C4

8_

01

_22

14

,8

-0,2

-1

,2

SC

48

_1_

24

13

,1

1,5

1

0,6

S

C4

8_

01

_18

14

,2

0,7

4

,8

AP

24

_1

_23

1

5,0

-0

,4

-2,4

SC

48

_1_

23

13

,1

1,5

1

0,5

A

P4

8_100_23

14

,0

0,7

4

,7

AP

24

_1

0_1

1

14

,2

1,6

1

0,4

A

P4

8_

1_15

1

4,3

0

,7

4,5


178

Supplementary Table S2 | Overview of affected genes in APEC phage resistant strains

Change

Change name

Affected gene

Annotation Function n Strain(s) Ref

Gene loss (P)

Group_25

neuE

NeuE (polysialic

acid biosynthesis protein)

E. coli K1 sialic

acid capsule synthesis

24

AP24_100_12,AP48_100_11, AP48_10_14, AP24_100_4, AP24_100_5, AP24_10_5,

SC24_100_7, SC24_100_12, SC24_100_13, SC24_10_23

SC24_1_8, SC24_01_11 SC24_01_15, SC48_100_8 SC48_10_8, SC48_10_10

SC48_1_10, SC48_1_11 SC48_1_13, SC48_1_24 SC48_01_9, SC48_01_12

SC24_01_5, SC24_100_6

[44,

54, 98, 99]

Gene

loss (P/C)

Group_

67

group_

67

Hypothetical

protein (1) Unknown

2

4

AP24_100_8, AP24_10_5 AP24_10_6, SC24_100_7 SC24_100_9, SC24_100_11

SC24_10_10, SC24_10_22 SC24_10_23, SC24_1_7 SC24_1_13, SC48_100_7

SC48_100_8, SC48_10_8 SC48_10_10, SC48_10_14

SC48_1_11, SC48_1_12 SC48_1_16, SC48_1_24 SC48_01_9, SC48_01_12

SC48_01_16, SC24_01_5

-

Gene loss (P/C)

Group_237

group_237

Hypothetical protein (2)

Unknown 10

AP24_100_5, AP48_1_24

SC24_100_6, SC24_1_8 SC24_100_11, SC48_01_9 SC48_1_11, SC48_1_12

SC48_100_12, SC48_100_8

-

Gene loss (P)

Group_310

group_310

Tetratricopeptide repeat (TPR) protein

Mediation of protein-protein interactions

8

SC24_10_23, SC24_01_11 SC48_100_8, SC48_10_10 SC48_10_23, SC48_1_10

SC48_1_24, SC24_100_6

[45]

SNP (AA/ST)

SNP 17/18/19

ackA Acetate kinase Phosphorylation of acetate to acetyl phosphate

7

AP24_10_6, SC24_100_7

SC24_100_9, SC24_1_17 SC48_100_7, SC48_100_8 SC48_1_22

[46]

SNP (s) SNP 1 SNP1* Hypothetical

protein (3) Unknown 6

AP24_100_4, AP24_100_5

AP24_10_5, AP24_10_6 SC24_100_6, SC24_01_5

-

Gene loss (P)

mprA_1

mprA_1

MarR family transcriptional regulator

Regulation of numerous cellular processes

5 AP24_10_5, SC24_100_7 SC24_10_10, SC24_10_23 SC48_1_11

[47]

Gene

loss (P/C)

rfbX wzx /

rfbX

Oligosaccharide flippase protein /

O-antigen transporter

Transport of O-

polysaccharides molecules

4 AP24_10_5, SC24_01_11

SC48_100_8, SC48_1_11 [48]

SNP (AA)

SNP 25 glyS Glycyl-tRNA synthetase beta

chain

tRNA recognition 4 SC24_100_10, SC24_100_11 SC24_100_12, SC48_100_12

[49]

Gene

loss (C) fadE

fadE /

yafH

Acyl-CoA

dehydrogenase

Dehydrogenation

of acyl-coenzymes A

3 SC24_1_8, SC24_1_9

SC48_1_9 [50]


179

Supplementary Table S2 | Continued

Change

Change name

Affected gene


Gene loss (C)

Group_306

yafV yafV / hydrolase family amidase

NAD(P)-binding

Metabolite repair enzyme

3 SC24_1_8, SC24_1_9 SC48_1_9

[51]

Gene loss (C)

ivy ivy

Vertebrate

lysosome inhibitor

Protection against

lysozyme-mediated cell wall hydrolysis

3 SC24_1_8, SC24_1_9 SC48_1_9

[52]

SNP (AA)

SNP 2 glnD Protein-PII uridylyltransferase (EC 2.7.7.59)

Nitrogen regulation

3 SC24_100_24, SC48_100_22 SC48_100_24

[53]

SNP

(ST) SNP 20 pta

Phosphate

acetyltransferase

Acetate

metabolism 3

SC24_10_10, SC48_10_10

SC48_10_11 [46]

Gene loss (P)

epsM_1

epsM_1

Acetyltransferase / NeuD protein

E. coli K1 sialic

acid capsule synthesis

2 AP24_10_5, AP24_100_8 [54, 99]

Gene loss (P)

Group_7

wzy

O1 family O-

antigen polymerase

Synthesis of the

LPS B-band O antigen

2 SC24_1_7, SC24_01_15 [48, 55]

SNP (ST/

AA)

SNP 6 / 7

ompA Outer membrane protein A (OmpA)

precursor

Key E. coli virulence factor

2 AP48_1_14, AP24_10_17 [56, 57]

SNP (AA)

SNP 29 / 30

relA

GTP

pyrophosphokinase

Synthesis of ppGpp from GTP

2 AP24_10_24, AP24_100_5 [58]

SNP

(AA)

SNP 12

/ 13 pykF

Pyruvate kinase

(PK)

Regulation of the glycolytic

pathway

1 SC24_1_23, SC24_10_16 [59]

Gene loss (P)

Group_10

wekM Glycosyltransferase family 4

Peptidoglycan biosynthesis

1 AP24_10_5 [48,87]

Gene

loss (P)

Group_

271

group_

271

Hypothetical

protein (4) Unknown 1 SC24_01_11 -

Gene

loss (P)

Group_

276 hokA

HokA (Type I TA

system toxin)

Toxic component

of HokA 1 SC24_10_23 [61]

Gene

loss (P)

Group_

333 ydfO

DUF1398 family

protein ydfO Unknown 1 AP24_10_5 [62]

Gene loss (P)

neuA neuA Acylneuraminate cytidylyltransferase

E. coli K1 sialic acid capsule synthesis

1 SC24_100_7 [54]

Gene loss (P)

splE splE Serine protease SplE

Various biological processes

1 SC24_100_7 [63]

SNP (s) SNP 10 dsbB

Periplasmic thiol:disulfide

oxidoreductase DsbB

Membrane-integrated protein

electron transfer catalyst

1 SC48_1_17 [64]

SNP

(AA) SNP 14 wzzB

O-antigen chain

length determinant protein WzzB

LPS biosynthesis 1 AP48_1_14 [65]

SNP (s) SNP 11 SNP11*

tRNA-Val-GAC Transfer of amino acids to the

ribosome

1 AP24_100_12 -


180

Supplementary Table S2 | Continued

Change

Change name

Affected gene


SNP

(AA) SNP 15 metG

Methionyl-tRNA

synthetase

Protein

biosynthesis 1 SC24_01_14 [66]

SNP (s) SNP 16 narP Nitrate/nitrite response regulator protein NarP

Gene expression regulation

1 SC24_01_16 [68]

SNP (AA)

SNP 21 gltX Glutamyl-tRNA synthetase GlnRS

Protein biosynthesis

1 AP24_10_5 [69]

SNP

(AA) SNP 22 hcp

T6SS component

Hcp

Bacterial

interaction with host cells

1 SC24_100_15 [70]

SNP (AA)

SNP 23 rapZ RNase adapter protein RapZ

Cell envelope

precursor sensing and signalling in

E. coli

1 AP48_100_23 [71–73]

SNP (s) SNP 24 rbsA Monosaccharide-transporting

ATPase

Transfer of solutes across

membranes

1 SC24_01_16 [74]

SNP (AA)

SNP 26 xylB Xylulose kinase

Phosphorylation of D-xylulose to D-xylulose 5-

phosphate

1 SC24_01_18 [75]

SNP (ST)

SNP 27 waaC

LPS core

heptosyltransferase I

LPS biosynthesis 1 SC24_100_6 [60]

SNP (ST)

SNP 28 waaY LPS core heptose (II) kinase RfaY

LPS biosynthesis 1 SC48_01_9 [60]

SNP (s) SNP 3 fur Ferric uptake regulation protein

FUR

Transcriptional regulation of iron

metabolism

1 SC48_01_17 [76]

SNP

(AA) SNP 31

SNP31

* 5'-nucleotidase

Hydrolyses the

phosphate group of 5′-nucleotides

1 SC48_01_17 [77]

SNP (AA)

SNP 32 phnD

Phosphonate ABC transporter substrate-binding

protein PhnD

Phosphonate uptake and utilisation

pathway

1 SC48_10_8 [78]

SNP (AA)

SNP 5 ybjT Uncharacterised protein YbjT

LPS biosynthesis and other core cell envelope

components

1 SC24_10_10 [79, 105

]

SNP (AA)

SNP 9 rne Ribonuclease E (RNase E)

RNA processing

and mRNA degradation

1 SC24_100_7 [80]

SNP (AA)

SNP 35 sgcR / yjhJ

sgc region transcriptional regulator

Transcriptional regulation

1 SC48_10_8 [81]

SNP

(AA) SNP 37

SNP37

*

Hypothetical

protein (5) - 1 SC48_10_22 -

SNP

SNP 4 /

8 / 33 / 34 / 36

- Non-coding

region -

1

*5

SC24_1_16 AP48_100_16

SC24_01_11 SC48_1_10 SC48_01_15

-

C = complete gene loss, P = partial gene loss, Point mutations: AA = missense, ST = nonsense, s = silent. * = gene not determined. Specified genetic change is included instead.


181

Su

pp

lem

en

tary

Ta

ble

S3

| O

verv

iew

of

ev

idence lev

el 1

-CR

ISP

R s

pacers

dete

cte

d in

ph

age-r

esi

stan

t st

rain

s

Sp

ace

r

#

Len

gth

(bp

) S

pace

r se

qu

ence

B

LA

ST

hit

G

enB

an

k

An

nota

tion

1

40

G

CG

CT

GC

GG

GT

CA

TT

TT

TG

AA

AT

TA

CC

CC

C

GC

TG

TG

CT

GT

E

sch

eri

ch

ia c

oli

str

ain

SB

02

58

h1

C

P0

71

95

4.1

G

en

era

l st

ress

pro

tein

2

54

G

CC

GT

TG

CC

GA

AT

GT

AG

GC

CG

GA

TA

AG

GC

GT

TC

AC

GC

CG

CA

TC

CG

GC

AA

CC

AG

C

Esc

heri

ch

ia c

oli

str

ain

EcP

F5

C

P0

54

23

6.1

-

3

34

C

TG

TA

AT

TT

TC

AT

GA

AA

GG

TG

GA

TG

GC

TG

C

GC

AC

Esc

heri

ch

ia c

oli

str

ain

CP

8-

3_

Sic

hu

an

pla

smid

pC

P8

-3-I

ncX

1

CP

05

37

40

.1

Pla

smid

(p

CP

8-3

-

IncX

1)

4

38

C

GG

AC

GC

AG

GA

TG

GT

GC

GT

TC

AA

TT

GG

AC

T

CG

AA

CC

AA

E

sch

eri

ch

ia c

oli

str

ain

LW

Y6

C

P0

72

20

4.1

tR

NA

-Va

l

5

58

G

GA

GC

CA

GA

AG

AA

CA

GA

TT

GA

TC

CG

CG

CA

AA

GC

CG

CC

GT

CG

AA

GC

TG

CT

AT

TG

CC

CG

T

Esc

heri

ch

ia c

oli

str

ain

SC

U-4

87

C

P0

54

45

4.1

Ele

ctr

on

tra

nsp

ort

co

mp

lex

su

bu

nit

Rsx

C

6

53

T

TT

CA

AG

TA

TT

GT

AA

AA

CA

TT

TG

AT

GC

AA

T

CG

CT

TA

TA

TT

GC

CG

AA

TC

TT

TT

G

Esc

heri

ch

ia p

ha

ge v

B_

Eco

M_G

29

M

K3

27

94

0.1

P

inA

pep

tid

ase

inh

ibit

or

7

24

G

GG

GG

GG

GG

GG

GG

GG

GG

GG

GT

TT

G

- -

-

8

27

C

CC

CC

CC

CC

CC

CC

CC

CC

CC

CC

CC

CC

CC

-

- -


182

3.4

Schematic overview of the experimental studies and main findings



183

The main findings of the experimental studies (Chapter 3.1, 3.2, and 3.3) are summarised in

Figure 1 and are discussed in the following pages (Chapter 4).

Figure 1 | Overview of the experimental studies (scientific chapters) and the main findings.

Chapter 4: General Discussion

185


Chapter 4

General Discussion

4.1 Avian pathogenic E. coli (APEC) - The need for alternative treatment options

APEC is one of the most important pathogens affecting poultry worldwide, with infections,

collectively referred to as colibacillosis, resulting in increased morbidity and mortality [1].

Various APEC serotypes have been associated with cases of colibacillosis, though, the three

serotypes O1, O2 and O78 account for most of the cases [1–3]. Antibiotics are commonly used

to control APEC infection in poultry, however the emergence of antimicrobial resistant

bacterial strains and antibiotic failures and restrictions have led to a growing interest in much-

needed complement treatment strategies, such as the prophylactic and therapeutic application

of phages (phage therapy) [1, 4–7]. Multiple studies evaluating the preventative and therapeutic

efficacy of phages against APEC in poultry have been conducted [reviewed by 1, 8]. These

studies suggest that phage application can be a valuable approach to prevent and control APEC

infections, and hereby, highly beneficial to both animals and humans in the framework of

animal welfare, farmers income, higher productivity, food security, and the environment as

well as reduced antimicrobial resistance. However, no phage treatment has yet advanced into


186

field application, possibly due to challenges in large-scale production and application as well

as regulatory hurdles [1, 8]. Considering the specific poultry sector economics, with large

numbers of animals with each a low intrinsic value and relatively short lifespan, phage

application (cost, route, etc.), like application of antibiotics, should be economically viable.

Due to the significant problem it represents to poultry welfare and industry, multi-drug resistant

APEC was chosen as the bacterial model organism in this work.

4.2 The remarkable diversity of E. coli-infecting phages

Phages are viruses that specifically infect bacteria. An opening statement such as “Phages are

the most abundant and diverse organisms on Earth” has frequently been used, when describing

phages [9–11]. In recent years, the understating of phage diversity has expanded enormously

through the increased availability of phage sequences [9, 12, 13]. Moreover, high-throughput

screening methods enable fast isolation and identification of hundreds of diverse phages [14].

However, the advancements have also revealed that we have only scratched the surface in the

discovery of novel viruses. The recent progress within the understanding of phage diversity is

also reflected in the extensive update of phage taxonomy in 2020 carried out by the ICTV [15].

The majority of phages isolated to date belong to the Caudovirales order of tailed double-

stranded DNA phages [16], possibly due to isolation biases and dedicated work of phage-

isolation programmes, such as the Science Education Alliance-Phage Hunters Advancing

Genomics and Evolutionary Science (SEA-PHAGES) [17–19]. In chapter 3.1, we investigated

the diversity of coliphages in the intestines of poultry as well as the overall coliphage genomic

diversity. Using the enrichment culture approach combined with WGS, we were able to isolate

and characterise Caudovirales coliphages belonging to the Myoviridae, Demerecviridae

(previously Siphoviridae), or the Drexlerviridae (previously Siphoviridae) families. The

classification was done by combining, morphological and WGS analysis (including whole

genome and phage marker gene analysis). We were able to isolate new phage species, some of

which belong to a newly created phage genus (Warwickvirus). Unravelling the enormous

diversity of phages and their genomes might be of great benefit. Exploring their diverse mode

of host interaction as well as structure-function correlation will contribute to research going on

to find new ways to efficiently exploit phages for application in various scientific and

therapeutic fields [16, 20]. Comparative genomics of phages allows for better understanding of

phage adhesion factors and the affinity between a bacterial receptor and a phage RBPs. Such


187

knowledge may be used to develop new adhesins, new ways to target bacterial pathogens, or

new ways to modulate the microbiome (such as probiotics) [21].

In this work we used E. coli K-12 derived laboratory strains (C600 and K514), as preliminary

studies (unpublished data) showed higher susceptibility compared to APEC strains for the

isolation of phages. Both laboratory strains were modified in their restriction, which most likely

promotes phage infection, and hereby, allows for isolation of more diverse phages [22]. Still,

one should be aware of the biases that the use of any strain creates and limits the possibility to

have a full picture of the coliphage diversity. Novel synthetic biology strategies may allow for

the creation of bacterial strains encoding pluripotent receptors but no phage resistance

mechanisms (R-M systems, Abi systems, CRISPR-Cas etc.), enabling phage isolation with

reduced bias. A metagenomic isolation bias-free approach is not suitable as individual phages

were needed for in vitro characterisation. However, for future work it would be interesting to

compare the findings of this current work with metagenomic analysis of the poultry phage

microbiome to determine the degree of diversity and evolution of coliphages that is reflected

in our collection.

4.2.1 Phage host spectrum

Phages differ in their host-specificity and show very diverse lytic behavior ranging from very

specifically lysing only one strain to infecting numerous strains [23]. In chapter 3.2, variation

in infectivity spectrum of tested coliphages was similarly observed with phages able to infect

all APEC strains while others not able to infect any of the tested APEC strains. The infectivity

of phages depends on their ability to successfully bind to the phage receptor of the host cells

and to overcome or evade strain-specific bacterial immunity [18, 24]. Caudovirales phages

bind to host surface receptors using phage RBPs, such as tail fibers, tail spikes or central tail

tips [22, 25]. Recently, Maffei et al. 2021 highlighted the patterns of phage receptor specificity

and showed that Siphoviridae and Myoviridae phages only target a very limited number of all

the E. coli OMPs. However, for most studied phages, including the ones used in this work, the

host receptor is unknown. RBPs are thought to be the primary determinant of phage host range,

and phages are believed to move between related hosts at least in part through genetic exchange

of or modifications in RBP-encoding genes [25–27]. For expanding the host spectrum, the

RBPs can also be engineered, based on extended sequence variation analysis and detailed

interaction studies, allowing to determine the importance of each of the RBPs in the adhesion


188

or go through a host co-evolutionary process [27–29]. Phage host range has been highlighted

as a crucial element for phage therapy application as broader/adapted infectivity, natural,

trained or engineered enable phages to be used against different target strains (thus different

phage receptor molecules) expanding the phage application spectrum [22, 27]. Also, the

different aspects of phage (extended) host spectrum (ability to infect vs. ability to replicate)

should be considered when choosing treatment strategy: “passive” treatment with initial

sufficient phage titer to reduce bacterial numbers (depends on phage bactericidal activity) or

“active” treatment requiring on-going in situ phage replication to reach/maintain numbers

sufficient to control the target bacteria [30, 31].

4.2.2 Hypothetical proteins of unknown function

Sequenced phage genomes are often annotated for gene identification and assignment of

putative gene function [32, 33]. As no single programme consistently outperforms the others,

most often multiple programmes are combined with manual interpretation of the finding to

achieve the highest accuracy [32]. In chapter 3.1, a similar approach combing automatic and

manual annotation was used. Still, in accordance with findings from various other studies, the

function of the majority of phage-encoded gene products are still not known and are referred

to as hypothetical proteins of unknown function [34–36]. Many phage genome annotations

include false positives (e.g., when every open reading frame longer than specific length is

annotated as a protein-coding gene) and/or false negatives (e.g., when genes shorter than 100

bp is falsely excluded as this is a commonly used cut-off length) [33]. Several approaches have

been put forth to overcome the major challenge of assigning function to hypothetical proteins,

including in silico structure/function prediction bioinformatic tools and comparative genomic

combined with functional genetic studies [35, 37, 38]. In this study, gene function of unknown

genes encoding hypothetical proteins was predicted using a comparative genomics approach,

comparing identified genes to homologue genes with defined functions in other related phage

genomes. Continuous advances in genome sequencing technologies, bioinformatics, and

proteomics will play a major role in unravelling the functions of hypothetical phage proteins

and provide a promising approach to identify more and novel targets of pathogenic bacteria

and to exploit phages therapeutic potential as well as their application in other scientific fields

[32, 34, 36].


189

4.3 Phage-host population growth dynamics

In the presence of sensitive bacteria, phage PK (phage movement and persistence in animal

body) and PD (phage host range and bacterial resistance) are fundamentally different from

those of antibiotics/chemical drugs as phages have the ability to self -replicate, co-evolve, and

elicit immune responses [31, 39–43]. Successful treatment is dependent on the suitable phages

reaching the targeted bacteria in sufficient numbers [31]. Accordingly, PK/PD are essential

parameters for better understanding the success of phage therapy [39, 44], and the sparse

knowledge is limiting its clinical applications [31, 41, 42].

Studies of in vitro phage-host interactions in liquid culture are commonly used to assess

bacterial growth dynamics in response to different phage exposure, using OD as input [45–49].

This method is fast and can be used for high-throughput comparison using 96-well plate

incubating spectrophotometers [50], similar to the experimental setup described in chapters

3.2 and 3.3. In our study, the growth dynamics of well-characterised virulent coliphages and

multidrug-resistant APEC strains at different MOIs were assessed. Using OD as input we

classified dynamic patters and determined the effect of phage species, APEC strain and MOI

on the observed dynamics. Out of the three variables the phage species had the most significant

effect of the dynamics outcome in the established phage-host model. Myoviridae phages

belonging to the Tevenvirinae subfamily (Tequatrovirus genus) exhibited the broadest host

range infecting both O1, O2, and O78 APEC strains. Similar findings have recently been

reported highlighting the great efficiency against target strains [22, 23, 51]. The specific

evolution of the phage-host population growth dynamics, including bacterial reduction and the

emergence of a phage-resistant bacterial population, depends on the various interactions of the

specific phage-host combination during the course of infection [52]. The binding affinity of the

phage to the bacterial host has been identified as one of the key parameters for the reduction of

the bacterial population size. Future studies looking at different phage and bacteria l gene

expression patterns might be able to identify additional crucial parameters related to the

different dynamics.

The co-evolution between phage and host bacterium has been described as antagonistic in

which a continuous arms race takes place [53–55]. However, similar to findings of Holguín et

al. (2019), no evidence of antagonistic co-evolution between phage and host strain was found

in the model used in chapter 3.2 and 3.3. Instead, an asymmetric dynamic in favour of the

bacteria was observed. Phages were not able to overcome existing resistance mechanisms nor

counteract phage resistance development. However, the (experimental) time factor could be of


190

great importance. Given enough time, co-cultures phages have the ability to evolve to

overcome the host phage resistance [57–59]. Phages applied therapeutically most likely do not

have a long time to adapt to overcome phage resistance mechanisms present in target bacteria.

Thus, potential adaptation of the phage(s) should happen prior to the application (using

engineering, training or other approaches described earlier). Some phages might be able to

overcome the resistance fast enough for direct application. Through studies with different

phage-host combinations, these fast-evolving phages could be identified along with the

underlying mechanisms conferring this ability.

In this work, the model employed to understand the coliphage-host population dynamics

involved in vitro co-cultured phage-host combination comprising a single phage (monophage)

and target strain. However, in practice the use of multiple phages (phage cocktail) is usually

favoured [60]. Nevertheless, the individual patterns may guide the composition of the cocktail

though the individual profiles are also important for the understanding of the dynamics. We

can potentially increase the host range, a more efficient suppression of bacterial growth can be

achieved, and phage resistance may be hampered [50, 51, 60, 61]. The inclusion of phage

cocktails as well as mixed cultures of target bacteria could help capture more complex

dynamics [62]. The model established in chapter 3.2 can be applied as an in vitro screening

approach for phage candidates for phage therapy, and may aid a more standardised and

quantitative evaluation [49]. The approach captures the ongoing phage-host population

dynamics and produces quantitative high-throughput data to determine phage host range, phage

virulence/infectivity, and emergence of bacterial phage-resistance, which facilitates a rational

design of phage therapies [50, 52]. However, in order to quantify and individually tract multiple

phages and bacteria in mixed cultures, alternative approaches such as real-time quantitative

PCR (RT-qPCR) might be needed [50, 63]. The effect of each defined dynamics pattern on the

course of infections needs to be determined in vivo.

4.4 Phage resistance in APEC

The ubiquity of phage predation has driven the evolution of various bacterial immune systems

targeting any step of the phage infection process [39, 64–66]. These mechanism include (but

are not limited to) R-M systems that target specific sequences on the invading phage genome,

Abi systems that lead to cell death or metabolic arrest of the bacterial cell upon phage infection,

and CRISPR-Cas, which provides acquired adaptive immunity through memorisation of past


191

phage attach [24, 67–69]. Moreover, new anti-phage strategies are still being discovered and

continues to expand the known arsenal of bacterial defence systems [70, 71].

One of the major hurdles to a successful phage therapy implementation is the presence and/or

emergence of phage-resistant bacteria [72, 73]. Whereas the first limits the applicability of

phage candidates for presumptive use and makes it harder to keep standardised phage cocktail

up to date, the latter may impede favourable treatment outcomes. Several in vitro studies have

reported the emergence of phage-resistant variants within a short period of time after phage

exposure [51, 73–75]. In our study, we detected bacterial growth after as early as ~7 hours with

exposure to a phage in chapter 3.2 and within 48 hours in the phage-host model of chapter

3.3. Contrary to more complex natural environments where mechanisms such as CRISPR-Cas

evolve rapidly [76, 77], in laboratory conditions, bacterial phage resistance is typically

conferred by mutations or loss of phage receptor-encoding genes, blocking phage adsorption

[76, 78]. Previously, E. coli knockout libraries, such as the Keio collection [79] (single-gene

knockout mutants of all non-essential genes of E. coli K-12) have been used to identify specific

genes involved in coliphage infection [22, 80]. However, this method can be laborious,

involving the screening of individual phage samples on very large numbers of mutant bacterial

strains if genes for phage receptors are unknown. Transposon mutant libraries have been used

to identify bacterial resistant mutants [81, 82]. While this approach is less laborious, it is limited

by the number of transposon mutants chosen for follow-up study, making it potentially

challenging to identify the receptors(s) used by a single phage [82]. Transposon mutagenesis

produces randomised gene mutations/insertions in the bacterial genome, not reflecting the

selective pressure conferred by the presence of phage. As we wanted to mimic natural

selections/patterns of targeted genes involved in phage resistance, natural mutation

development was used. This approach, however, necessitates comparing WGS of phage -

resistant mutants with a well-annotated reference genome to identify mutations. Spontaneous

mutation(s) can, but are not guaranteed to, arise in cell surface structures that are required for

initial phage adsorption to bacteria [78]. These mutations include both one or more single point

mutation(s) as well as partial and/or complete gene loss. In this study, gene loss and/or

mutations in genes encoding the OmpA and LPS were detected in the phage-resistant variants

and that alone was sufficient for bacterial resistance. However, in accordance with previous

studies we also found that multiple linked mutations may be necessary for full phage resistance

[78, 82, 83]. In some of our phage-resistant mutant strains from chapter 3.3 we detected point

mutations and/or gene loss in two or three genes. Affected genes included genes known to be


192

involved in phage resistance as well as genes not previously linked to resistance, indicating

that the phage-host interaction might be more complex than previously thought. The latter

group included, among others, genes encoding hypothetical proteins with unknow function.

However, the potential involvement in phage resistance remains to be confirmed. Further

characterisation of the hypothetical proteins, including gene function assignment, will also be

necessary to fully understand the involvement of this genetic changes in phage resistance.

Some phages have been shown to be more able to circumvent the bacterial resistance, and the

mechanisms they use for this may help overcome these resistance mechanisms [58, 84]. It has

been shown that (Myoviridae) Tevenvirinae phages exhibited broad resistance to R-M systems

due to cytosines modification in their genomes [22, 85]. Maffei et al. (2021) found

(Myoviridae) Vequintavirinae phages to have an exceptional host range, possibly due to their

ability to bypass the O-antigen barrier. However, they also reported a high sensitivity to various

R-M systems. Moreover, (Myoviridae) Ounavirinae phages belonging to the Felixounavirus

genus showed remarkable sensitivity to several tested R-M systems. Engineering these R-M-

sensitive phages by inserting the cytosine modification mechanisms may prevent this

resistance, however engineering of phages is still problematic [86], but technologies are

evolving and may become available in the future. This would lead to engineered phages with

a wanted host spectrum and less sensitivity towards bacterial phage resistance mechanisms. In

our study, we found the Tevenvirinae phages able to circumvent these natural immune systems

and infect most of the O1, O2, and O78 APEC strains. Very recent, Maffei et al. (2021) showed

a correlation between Drexlerviridae (previously Siphoviridae) phylogeny and

sensitivity/resistance to bacterial immunity, though the mechanisms behind still need to be

unravelled. However, while it might be an interesting target for phage engineering, high

resistance to bacterial immunity is not a guarantee for successful treatment outcome (chapter

3.2). Another hurdle to be considered is the fact that these engineered phages fall under the

regulations of genetically modified organisms (GMO). Such phages can be subjected to

additional regulations compared to natural phages and this may cause problems for in filed

acceptance of the therapies [87]. The safety and consequences of such genetically engineered

phages should be well determined [73].

Understanding the structural and molecular mechanisms of phage-host interaction is thus

crucial for the application of phages [25]. The use of phage cocktails, composed of phages

using different host receptors, has proved to be advantageous against several bacterial

pathogens and less likely to select for phage-resistant strains [51, 88]. Though it remains no


193

guarantee for successful application [51, 89–91]. In this work only single phage single host-

combinations were used when studying growth dynamics (chapter 3.2) and phage resistance

development (chapter 3.3). In future experiments, carefully selected phages (see below), based

on the single reactions, can be combined in a cocktail, and exposed to target APEC strains.

However, more work is necessary to define what properties would be necessary in a good

cocktail. Our work contributed to the variables of interest in composing such cocktail.

4.4.1 The cost of phage resistance

Emergence of phage-resistant bacteria is, with our current knowledge and capacities to

engineer phages, unavoidable, and resistance confers a clear selective advantage of the

bacterium in the presence of phage. However, the resistance often comes with a cost for the

bacterium as demonstrated in our results, and the magnitude of the trade-off depends on the

genetic basis of the resistance as well as the environmental context [31, 39, 66, 72, 78, 92].

Immune systems such as R-M systems and CRISPR-Cas is associated with fitness cost as these

can be extremely energy consuming and may use resources that would otherwise be invested

in cell growth [66, 93, 94]. Resistance conferred by modifications to surface LPS, membrane

porins, siderophores, efflux pumps, pili, and flagella may come with fitness cost as well,

including reduced virulence, decreased resistance to environmental pressures, colonisation

defects, reduced growth rate, reduced motility, and re-sensitisation to antibiotics or the host

immune system [51, 66, 78, 95–98]. The cost of losing surface receptors, is likely to be lower

in laboratory settings, where nutrients are readily available and competition is less, than in

natural environments [66, 78, 99]. Depending on the risk/frequency of phage infection, either

CRISPR systems (low risk) or surface modification (high risk) is favoured as resistance

mechanisms by the bacterium [66, 100]. Inducible resistance mechanisms (triggered upon

phage infection) such as CRISPR is associated with an induced cost of resistance, whereas

constitutive mechanisms (always active), such as loss of cell surface receptors, are associated

with a fixed cost. The overall cost of the inducible resistance will depend on the infection

frequency and will determine the most favourable defence strategy. The phage-resistant

variants obtained in chapter 3.3 did not show changes their CRISPR loci, however, several

genes associated with phage adoption were affected by point mutation or gene loss.

Interestingly, the greatest fitness cost, measured by decrease in growth relative to the

susceptible WT strain, was conferred by mutation in genes with unknown function. These

genes can serve as potential targets for new therapeutic applications. By inhibiting the function


194

of these genes using for example chemicals, the growth of these bacteria can be inhibited, and

the infection can be overcome by the susceptible host. Deletions or mutations in these genes

can also be used in vaccine developments if shown that these reduce the virulence to an

acceptable level.

While phages, antibiotics, or animal host immunity may not be sufficient for clearing bacterial

infections singlehandedly, combination therapies with phage(s) and antibiotics are clinically

promising as they are showing synergistic effects [88, 101–105]. Phages can utilise critical

bacterial surface molecules that provide either defence from antibiotics via efflux, uptake of

nutrients in resource-limited host environments, or general cell wall maintenance [78].

Accordingly, by understanding and exploiting phage resistance, phage therapy can “steer”

pathogenic bacteria toward deleterious surface mutations that allow for more favourable

treatment outcomes and extend the useful lifespan of antibiotics that otherwise would have

been discarded [95, 104]. Since not all mutations in surface factors will lead to desirable

treatment outcome, and given the complexity of phage–bacteria interactions, special

considerations to context-dependent (in vitro vs. in vivo) fitness cost are required when

developing effective phage steering therapies for clinical use [39, 78, 95].

4.5 Acquisition and selection of suitable phages for phage therapy

Phages can readily be isolated from various environments, and a much larger set of phages is

often available than can be used in any application [10, 18, 106]. When excess phages are

available, how to select the best phage(s)? Before administration, phage therapy candidate

phages undergo careful examination of their safety and interaction with target bacteria [18],

currently best done by WGS and determination of the host range and dynamics. Recently, a

framework for a standardised in vitro evaluation of phage candidates for phage therapy was

suggested [49]. A key first step is to develop a large collection of well-characterised candidate

phages followed by matching of phage(s) to target bacteria, including an investigation of the

phage-host growth kinetics in liquid cultures. Subsequently, the interactions between target

bacteria and phage(s) should be examined as well as the effect of the host immune system on

the phage(s) to obtain optimal synergistic results [20]. Seen the current difficulties in

engineering phages, such phages are not yet available, but may represent another solution to

the therapeutic use of phages as discussed above.


195

In chapter 3.1, a well-characterised collection of virulent coliphages was established. All

phage comprised genetic properties desirable for phage therapy, including no genes encoding

any known toxins, virulence factors, or resistance to antibiotics [18, 107]. However, when in

vitro phage-host growth kinetics were determined (chapter 3.2), half of the phages were not

able to infect any of the target bacteria (O1, O2, and O78 APEC strains). One possible

explanation for lack of infectivity might be found in the choice of bacterial strain(s) used to

isolate the phages. In this work, phages were initially isolated on non-pathogenic E. coli

laboratory strains. We hypothesised that these would provide a broad diversity of phages as

opposed to WT AEPC strains, which most likely are better equipped to resist a wide array of

phage infections. For phage therapy application, the phage lytic potential on relevant

pathogenic bacterial strains is needed to determine their clinical coverage and importance for

inclusion in phage cocktails [108]. Using multiple host strains during isolation can more

reliably select for broader host range phages (able to inf ect multiple bacterial species),

compared to phages isolated using a single bacterial host strain [109]. In accordance with

previous findings [51, 56, 106, 110], we found the phage species, the APEC strain, and the

MOI all significantly influenced the observed dynamics, of which the phage species factor had

the most significant influence. Bull and Gill (2014) recommended a comparative approach to

predict efficacy. By comparing phages with differences in their dynamic properties and

treatment efficacy, properties associated with success can be identified. When combining

multiple phages in a well composed cocktail, it is possible to obtain a synergistic effect against

target bacteria, including faster killing and/or delay/hindrance of resistance development [51,

56, 61, 101, 111]. However, Holguín et al. (2019) and Pinto et al. (2021) showed that the

combination of any two phages may not have the desired effect. Selecting suitable phage(s) for

therapeutic application require sufficient understanding of phage-host interaction and can be

the difference between application success or failure. Still, a high in vitro efficiency is not

necessarily able to simulate the complex in vivo interplay of phage, bacteria and animal host,

and therefore, not a guarantee for successful therapy application [51, 89–91].

The use of standard growth conditions (LB medium, 37 °C, regular aeration), like for most

phage work of the last decades, was applied in this work to generate systematic, reproducible

data. However, such conditions mimic only a small part (if any) of the diverse environments

and physiological conditions that add up to the complex habitats of E. coli and its phages.

Safety and efficacy data obtained from preliminary in vitro testing of host range, efficiency of

plating, killing curves, phage kinetics, phage receptor(s) determination, cocktail optimisation


196

and phage/antibiotic interactions can be combined with data obtained from complex ex vivo

infection models mimicking the microenvironment [108, 112]. Also, stability data of the phage

(cocktail) at various temperatures, pH ranges, and oxygen and nutrient levels, mimicking both

the environment of the animal host and the application route (feed, water, injection, spray), is

essential to determine the potential for therapeutic use as well as shelf -life expectancy [113].

These steps will provide robust preclinical data to support the translation of in vitro data to in

vivo application. Before in vivo application, candidate phage (cocktail) with most desirable in

vitro properties, should be produced at high titre and meet regulatory criteria, purified form

toxins and produced under GMP conditions [108]. Companies may however be reluctant to

carry out large-scale production, until overall cost, including infrastructure and equipment

costs, have decreased enough to make it profitable [114]. Several complex external factors may

influence the application success in vivo, such as rapid clearance of the phages by the animal

host immune system and interactions of phages with other microorganisms [108, 113]. Future

evaluation of phages in vivo will include steps to determine phage safety and [108, 115]. Safety

data should provide information on the impact on other components of the poultry microbiome

and the eventual immunopathological responses. Efficacy data should provide information on

the phages impact on the colonisation of the target bacteria and on the curing effect.

The increase of sequence data of phages, and the improvement of sequence analysis tools as

well as the increasing knowledge on phage properties and phage-host interactions will aid a

more refined translation of in vitro data to in vivo outcomes in future phage therapy

applications.

4.6 Conclusions and future perspectives

This thesis contributes to a better understanding of the coliphage diversity as well as the phage -

host interaction and population dynamics. First, we provided new insight into the diversity of

coliphages in the intestines of poultry, where they live together with the target APEC, and have

established a well-characterised collection of lytic coliphages, some of which could be

candidates for therapeutic application against APEC infections. The coliphage collection can

advantageously be extended, including phages from currently unrepresented genera, such as

Vequintavirinae. Secondly, a growth dynamics in vitro model was combined with two data

mining techniques and quantitative scoring algorithms. This methodology can be used to

rapidly screen for novel phage candidates as part of the initial steps of phage therapy


197

development. Moreover, future research using this methodology could benefit from linking

specific phage and/or bacterial properties to specific growth dynamics and therapeutic

outcomes. Finally, we identified underlying mechanisms of in vitro phage resistance in APEC

by identifying genes affected by genetic change(s). Among others, these genes included two

genes encoding hypothetical proteins with unknown function, which have never been described

as involved in phage resistance. Considering the prevalence of genetic change(s), the great

associated fitness cost as well as possible involvement in phage-resistance, these genes may be

promising targets for future investigations and targets for therapies against APEC. Finally, the

therapeutic potential of multiple phages (phage cocktail) should be determined and potential

synergistic effects with antibiotics should be explored in vitro as well as in vivo.

The process form phage isolation to successful therapeutic application comprises many steps.

It is clear that currently, we are not able to translate well the in silico to the in vitro data, though

progress is made in the essential first steps in this thesis, namely characterising candidate

phages and their interactions with target pathogenic bacteria. There is still a large need to

extend this work. The phage adhesins need to be better characterised as well as the relevance

of the mutations found in target APEC strains. However, if enough phage receptors are

identified for sequenced phages, bioinformatic approaches might be extended to allow for high-

throughput identification of the phage receptor(s) and bacterial host(s) based on the genomes

of the phage alone. Similarly, essential knowledge on the phage-host interactions may be

obtained using machine learning models and be applied to predict PK/PD in suitable in vivo

models. Continuous model refinement can be applied if the experimental data do not fit the

predicted ones. With increasing well-characterised phage collections housed in reference phage

banks that can rapidly be matched with target bacterium, knowledge gaps in phage research are

being filled and continuous advanced in the field of phage research facilitate future effective

translation into promising therapeutic application.


198

References




[2] Huja, S.; Oren, Y.; Trost, E.; Brzuszkiewicz, E.; Biran, D.; Blom, J.; Goesmann, A.; Gottschalk, G.;

Hacker, J.; Ron, E. Z.; et al. Genomic Avenue to Avian Colisepticemia. MBio, 2015, 6 (1), 1–13.






[4] Wernicki, A.; Nowaczek, A.; Urban-Chmiel, R. Bacteriophage Therapy to Combat Bacterial

Infections in Poultry. Virol. J., 2017, 14 (1), 179. https://doi.org/10.1186/s12985-017-0849-7.

[5] Tang, K. L.; Caffrey, N. P.; Nóbrega, D. B.; Cork, S. C.; Ronksley, P. E.; Barkema, H. W.; Polachek,

A. J.; Ganshorn, H.; Sharma, N.; Kellner, J. D.; et al. Restricting the Use of Antibiotics in Food -

Producing Animals and Its Associations with Antibiotic Resistance in Food-Producing Animals and

Human Beings: A Systematic Review and Meta -Analysis. Lancet Planet. Heal., 2017, 1 (8), e316–

e327. https://doi.org/10.1016/S2542-5196(17)30141-9.




[7] Brives, C.; Pourraz, J. Phage Therapy as a Potential Solution in the Fight against AMR: Obstacles

and Possible Futures. Palgrave Commun., 2020, 6, 100.

[8] Żbikowska, K.; Michalczuk, M.; Dolka, B. The Use of Bacteriophages in the Poultry Industry.

Animals, 2020, 10 (5), 872. https://doi.org/10.3390/ani10050872.

[9] Dion, M. B.; Oechslin, F.; Moineau, S. Phage Diversity, Genomics and Phylogeny. Nat. Rev.

Microbiol., 2020, 12. https://doi.org/10.1038/s41579-019-0311-5.







[12] Hayes, S.; Mahony, J.; Nauta, A.; Van Sinderen, D. Metagenomic Approaches to Assess

Bacteriophages in Various Environmental Niches. Viruses, 2017, 9 (6), 127.

https://doi.org/10.3390/v9060127.

[13] Olsen, N. S.; Kot, W.; Junco, L. M. F.; Hansen, L. H. Exploring the Remarkable Diversity of

Escherichia Coli Phages in the Danish Wastewater Environment, Including 91 Novel Phage Species.

bioRxiv, 2020.

[14] Olsen, N. S.; Hendriksen, N. B.; Hansen, L. H.; Kot, W. A New High-Throughput Screening Method

for Phages: Enabling Crude Isolation and Fast Identification of Diverse Phages with Therapeutic

Potential. Phage, 2020, 1 (3), 137–148. https://doi.org/10.1089/phage.2020.0016.

[15] Krupovic, M.; Turner, D.; Morozova, V.; Dyall, M.; Hanna, S.; Edwards, R.; Dutilh, B. E.; Lehman,

S. M.; Reyes, A.; Baquero, D. P. Bacterial Viruses Subcommittee and Archaeal Viruses

Subcommittee of the ICTV : Update of Taxonomy Changes in 2021. Arch. Virol., 2021, 1–6.

https://doi.org/10.1007/s00705-021-05205-9.


199

[16] Zrelovs, N.; Dislers, A.; Kazaks, A. Motley Crew: Overview of the Currently Available Phage

Diversity. Front. Microbiol., 2020, 11. https://doi.org/10.3389/fmicb.2020.579452.





[18] Hyman, P. Phages for Phage Therapy: Isolation, Characterization, and Host Range Breadth.

Pharmaceuticals, 2019, 12 (1), 35. https://doi.org/10.3390/ph12010035.

[19] Cook, R.; Brown, N.; Redgwell, T.; Rihtman, B.; Barnes, M.; Clokie, M.; Stekel, D. J.; Hobman, J.;

Jones, M. A.; Millard, A. INfrastructure for a PHAge REference Database: Identification of Large-

Scale Biases in the Current Collection of Cultured Phage Genomes. Phage, 2021, 1–10.


[20] Rohde, C.; Wittmann, J. Phage Diversity for Research and Application. Antibiotics, 2020, 9 (11),

734. https://doi.org/10.3390/antibiotics9110734.

[21] Szymczak, P.; Rau, M. H.; Monteiro, J. M.; Pinho, M. G.; Filipe, S. R.; Vogensen, F. K.; Zeidan, A.

A.; Janzen, T. A Comparative Genomics Approach for Identifying Host -Range Determinants in

Streptococcus Thermophilus Bacteriophages. Sci. Rep., 2019, 9, 7991.

https://doi.org/10.1038/s41598-019-44481-z.





[23] Korf, I. H. E.; Meier-Kolthoff, J. P.; Adriaenssens, E. M.; Kropinski, A. M.; Nimtz, M.; Rohde, M.;

van Raaij, M. J.; Wittmann, J. Still Something to Discover: Novel Insights into Escherichia Coli

Phage Diversity and Taxonomy. Viruses, 2019, 11 (5), 1–29. https://doi.org/10.3390/v11050454.

[24] Hyman, P.; Abedon, S. T. Bacteriophage Host Range and Bacterial Resistance. In Advances in

Applied Microbiolog; Elsevier Inc., 2010; Vol. 70, pp 217–248. https://doi.org/10.1016/S0065-

2164(10)70007-1.

[25] Nobrega, F. L.; Vlot, M.; de Jonge, P. A.; Dreesens, L. L.; Beaumont, H. J. E.; Lavigne, R.; Dutilh,

B. E.; Brouns, S. J. J. Targeting Mechanisms of Tailed Bacteriophages. Nat. Rev. Microbiol., 2018,

16 (12), 760–773. https://doi.org/10.1038/s41579-018-0070-8.




[27] Abdelsattar, A.; Dawooud, A.; Rezk, N.; Makky, S.; Safwat, A.; Richards, P.; El-Shibiny, A. How

to Train Your Phage: The Recent Efforts in Phage Training. Biologics, 2021, 1 (2), 70–88.

https://doi.org/10.3390/biologics1020005.



https://doi.org/10.1038/s41467-017-00158-7.

[29] Yehl, K.; Lemire, S.; Yang, A. C.; Ando, H.; Mimee, M.; Torres, M. D. T.; de la Fuente-Nunez, C.;

Lu, T. K. Engineering Phage Host-Range and Suppressing Bacterial Resistance through Phage Tail

Fiber Mutagenesis. Cell, 2019, 179 (2), 459-469.e9. https://doi.org/10.1016/j.cell.2019.09.015.

[30] Payne, R. J. H.; Jansen, V. A. A. Pharmacokinetic Principles of Bacteriophage Therapy. Clin.

Pharmacokinet., 2003, 42 (4), 315–325. https://doi.org/10.2165/00003088-200342040-00002.

[31] Dąbrowska, K.; Abedon, S. T. Pharmacologically Aware Phage Therapy: Pharmacodynamic and

Pharmacokinetic Obstacles to Phage Antibacterial Action in Animal and Human Bodies. Microbiol.


200

Mol. Biol. Rev., 2019, 83 (4), e00012-19. https://doi.org/10.1128/mmbr.00012-19.

[32] Lazeroff, M.; Ryder, G.; Harris, S. L.; Tsourkas, P. K. Phage Commander, an Application for Rapid

Gene Identification in Bacteriophage Genomes Using Multiple Programs. PHAGE, 2021, 00 (00),

1–10. https://doi.org/10.1089/phage.2020.0044.

[33] Mcnair, K.; Zhou, C.; Dinsdale, E. A.; Souza, B.; Edwards, R. A. PHANOTATE: A Novel Approach

to Gene Identification in Phage Genomes. Bioinformatics, 2019, 35 (22), 4537–4542.

https://doi.org/10.1093/bioinformatics/btz265.

[34] Mohanraj, U.; Wan, X.; Spruit, C. M.; Skurnik, M.; Pajunen, M. I. A Toxicity Screening Approach

to Identify Bacteriophage-Encoded Anti-Microbial Proteins. Viruses, 2019, 11 (11), 1057.

https://doi.org/10.3390/v11111057.

[35] Klumpp, J.; Fouts, D. E.; Sozhamannan, S. Bacteriophage Functional Genomics and Its Role in

Bacterial Pathogen Detection. Brief. Funct. Genomics, 2013, 12 (4), 354–365.

https://doi.org/10.1093/bfgp/elt009.

[36] Kasurinen, J.; Spruit, C. M.; Wicklund, A.; Pajunen, M. I.; Skurnik, M. Screening of Bacteriophage

Encoded Toxic Proteins with a Next Generation Sequencing-Based Assay. Viruses, 2021, 13 (5),

750. https://doi.org/10.3390/v13050750.

[37] Ijaq, J.; Malik, G.; Kumar, A.; Das, P. S.; Meena, N.; Bethi, N.; Sundararajan, V. S.; Suravajhala, P.

A Model to Predict the Function of Hypothetical Proteins through a Nine-Point Classification

Scoring Schema. BMC Bioinformatics, 2019, 20 (1), 14. https://doi.org/10.1186/s12859-018-2554-

y.

[38] Zhou, C. L. E.; Kimbrel, J.; Edwards, R.; McNair, K.; Souza, B. A.; Malfatti, S. MultiPhATE2: Code

for Functional Annotation and Comparison of Phage Genomes. G3 Genes, Genomes, Genet., 2021,

11 (5), jkab074. https://doi.org/10.1093/g3journal/jkab074.

[39] Dąbrowska, K. Phage Therapy: What Factors Shape Phage Pharmacokinetics and Bioavailability?

Systematic and Critical Review. Med. Res. Rev., 2019, 39 (5), 2000–2025.

https://doi.org/10.1002/med.21572.

[40] Qadir, M. I.; Mobeen, T.; Masood, A. Phage Therapy: Progress in Pharmacokinetics. Brazilian J.

Pharm. Sci., 2018, 54 (1), e17093. https://doi.org/10.1590/s2175-97902018000117093.

[41] Dhungana, G.; Nepal, R.; Regmi, M.; Malla, R. Pharmacokinetics and Pharmacodynamics of a Novel

Virulent Klebsiella Phage Kp_Pokalde_002 in a Mouse Model. Front. Cell. Infect. Microbiol., 2021,

11, 684704. https://doi.org/10.3389/fcimb.2021.684704.

[42] Lin, Y. W.; Chang, R. Y.; Rao, G. G.; Jermain, B.; Han, M. L.; Zhao, J. X.; Chen, K.; Wang, J. P.;

Barr, J. J.; Schooley, R. T.; et al. Pharmacokinetics/Pharmacodynamics of Antipseudomonal

Bacteriophage Therapy in Rats: A Proof-of-Concept Study. Clin. Microbiol. Infect., 2020, 26 (9),

1229–1235. https://doi.org/10.1016/j.cmi.2020.04.039.





[45] Konopacki, M.; Grygorcewicz, B.; Dołęgowska, B.; Kordas, M.; Rakoczy, R. PhageScore: A Simple

Method for Comparative Evaluation of Bacteriophages Lytic Activity. Biochem. Eng. J., 2020, 161,

107652. https://doi.org/10.1016/j.bej.2020.107652.







201

https://doi.org/10.3390/v10040189.




[49] Gelman, D.; Yerushalmy, O.; Alkalay-Oren, S.; Rakov, C.; Ben-Porat, S.; Khalifa, L.; Adler, K.;

Abdalrhman, M.; Coppenhagen-Glazer, S.; Aslam, S.; et al. Clinical Phage Microbiology: A

Suggested Framework and Recommendations for the in-Vitro Matching Steps of Phage Therapy.

The Lancet Microbe, 2021, 2 (10), e555–e563. https://doi.org/10.1016/S2666-5247(21)00127-0.

[50] Gayder, S.; Parcey, M.; Nesbitt, D.; Castle, A. J.; Svircev, A. M. Population Dynamics between

Erwinia Amylovora, Pantoea Agglomerans and Bacteriophages: Exploiting Synergy and

Competition to Improve Phage Cocktail Efficacy. Microorganisms, 2020, 8 (9), 1449.

https://doi.org/10.3390/microorganisms8091449.

[51] Korf, I. H. E.; Kittler, S.; Bierbrodt, A.; Mengden, R.; Rohde, C.; Rohde, M.; Kroj, A.; Lehnherr, T.;

Fruth, A.; Flieger, A.; et al. In Vitro Evaluation of a Phage Cocktail Controlling Infect ions with

Escherichia Coli. Viruses, 2020, 12 (12), 1470. https://doi.org/10.3390/v12121470.

[52] Loessner, H.; Schlattmeier, I.; Anders-Maurer, M.; Bekeredjian-Ding, I.; Rohde, C.; Wittmann, J.;

Pokalyuk, C.; Krut, O.; Kamp, C. Kinetic Fingerprinting Links Bacteria-Phage Interactions with

Emergent Dynamics: Rapid Depletion of Klebsiella Pneumoniae Indicates Phage Synergy.

Antibiotics, 2020, 9 (7), 408. https://doi.org/10.3390/antibiotics9070408.

[53] Buckling, A.; Rainey, P. B. Antagonistic Coevolution between a Bacterium and a Bacteriophage.


[54] Poullain, V.; Gandon, S.; Brockhurst, M. A.; Buckling, A.; Hochberg, M. E. The Evolution of

Specificity in Evolving and Coevolving Antagonistic Interactions between a Bacteria and Its Phage.

Evolution (N. Y)., 2008, 62 (1), 1–11. https://doi.org/10.1111/j.1558-5646.2007.00260.x.

[55] Perry, E. B.; Barrick, J. E.; Bohannan, B. J. M. The Molecular and Genetic Basis of Repeatable

Coevolution between Escherichia Coli and Bacteriophage T3 in a Laboratory Microcosm. PLoS One,

2015, 10 (6), e0130639. https://doi.org/10.1371/journal.pone.0130639.

[56] Holguín, A.; Cárdenas, P.; Prada-Peñaranda, C.; Rabelo Leite, L.; Buitrago, C.; Clavijo, V.; Oliveira,

G.; Leekitcharoenphon, P.; Møller Aarestrup, F.; Vives, M. Host Resistance, Genomics and

Population Dynamics in a Salmonella Enteritidis and Phage System. Viruses, 2019, 11 (2), 188.

https://doi.org/10.3390/v11020188.

[57] Meyer, J. R.; Dobias, D. T.; Weitz, J. S.; Barrick, J. E.; Quick, R. T.; Lenski, R. E. Repeatability and

Contingency in the Evolution of a Key Innovation in Phage Lambda. Science (80-. )., 2012, 335

(6067), 428–432. https://doi.org/10.1126/science.1214449.

[58] Wandro, S.; Oliver, A.; Gallagher, T.; Weihe, C.; England, W.; Martiny, J. B. H.; Whiteson, K.

Predictable Molecular Adaptation of Coevolving Enterococcus Faecium and Lytic Phage EfV12 -

Phi1. Front. Microbiol., 2019, 9, 3192. https://doi.org/10.3389/fmicb.2018.03192.

[59] Hall, A. R.; Scanlan, P. D.; Buckling, A. Bacteria‐phage Coevolution and the Emergence of

Generalist Pathogens. Am. Nat., 2011, 177 (1), 44–53. https://doi.org/10.1086/657441.

[60] Chan, B. K.; Abedon, S. T.; Loc-Carrillo, C. Phage Cocktails and the Future of Phage Therapy.

Future Microbiol., 2013, 8 (6), 769–783. https://doi.org/10.2217/fmb.13.47.

[61] Naghizadeh, M.; Torshizi, M. A. K.; Rahimi, S.; Dalgaard, T. S. Synergistic Effect of Phage Therapy

Using a Cocktail Rather than a Single Phage in the Control of Severe Colibacillosis in Quails. Poult.

Sci., 2019, 98 (2), 653–663. https://doi.org/10.3382/ps/pey414.

[62] Betts, A.; Gifford, D. R.; MacLean, R. C.; King, K. C. Parasite Diversity Drives Rapid Host

Dynamics and Evolution of Resistance in a Bacteria-Phage System. Evolution (N. Y)., 2016, 70 (5),

969–978. https://doi.org/10.1111/evo.12909.


202

[63] Duyvejonck, H.; Merabishvili, M.; Pirnay, J.; Vos, D. De; Simaey, L. Van; Vermeulen, S.; Mechelen,

E. Van; Vaneechoutte, M. Development of a QPCR Platform for Quantification of the Five

Bacteriophages Within. Sci. Rep., 2019, 9, 13893. https://doi.org/10.1038/s41598-019-50461-0.

[64] Hampton, H. G.; Watson, B. N. J.; Fineran, P. C. The Arms Race between Bacteria and Their Phage

Foes. Nature, 2020, 577 (7790), 327–336. https://doi.org/10.1038/s41586-019-1894-8.

[65] Dy, R. L.; Richter, C.; Salmond, G. P. C.; Fineran, P. C. Remarkable Mechanisms in Microbes to

Resist Phage Infections. Annu. Rev. Virol., 2014, 1 (1), 307–331. https://doi.org/10.1146/annurev-

virology-031413-085500.

[66] van Houte, S.; Buckling, A.; Westra, E. R. Evolutionary Ecology of Prokaryotic Immune

Mechanisms. Microbiol. Mol. Biol. Rev., 2016, 80 (3), 745–763.


[67] Lopatina, A.; Tal, N.; Sorek, R. Abortive Infection: Bacterial Suicide as an Antiviral Immune

Strategy. Annu. Rev. Virol., 2020, 7 (1), 371–384. https://doi.org/10.1146/annurev-virology-011620-

040628.

[68] Azam, A. H.; Tanji, Y. Bacteriophage-Host Arm Race: An Update on the Mechanism of Phage

Resistance in Bacteria and Revenge of the Phage with the Perspective for Phage Therapy. Appl.

Microbiol. Biotechnol., 2019, 103, 2121–2131. https://doi.org/10.1007/s00253-019-09629-x.




[70] Picton, D. M.; Luyten, Y. A.; Morgan, R. D.; Nelson, A.; Smith, D. L.; Dryden, D. T. F.; Hinton, J.

C. D.; Blower, T. R. The Phage Defence Island of a Multidrug Resistant Plasmid Uses Both BREX

and Type IV Restriction for Complementary Protection from Viruses. Nucleic Acids Res., 2021,

gkab906. https://doi.org/10.1093/nar/gkab906.

[71] Doron, S.; Melamed, S.; Ofir, G.; Leavitt, A.; Lopatina, A.; Keren, M.; Amitai, G.; Sorek, R.

Systematic Discovery of Antiphage Defense Systems in the Microbial Pangenome. Science (80-. ).,

2018, 359 (6379), eaar4120. https://doi.org/10.1126/science.aar4120.



[73] Pires, D. P.; Costa, A. R.; Pinto, G.; Meneses, L.; Azeredo, J. Current Challenges and Future

Opportunities of Phage Therapy. FEMS Microbiol. Rev., 2020, fuaa017, 1–17.

https://doi.org/10.1093/femsre/fuaa017.

[74] Oechslin, F.; Piccardi, P.; Mancini, S.; Gabard, J.; Moreillon, P.; Entenza, J. M.; Resch, G.; Que, Y.

A. Synergistic Interaction between Phage Therapy and Antibiotics Clears Pseudomonas Aeruginosa

Infection in Endocarditis and Reduces Virulence. J. Infect. Dis., 2017, 215 (5), 703–712.

https://doi.org/10.1093/infdis/jiw632.



[76] Alseth, E. O.; Pursey, E.; Luján, A. M.; McLeod, I.; Rollie, C.; Westra, E. R. Bacterial Biodiversity

Drives the Evolution of CRISPR-Based Phage Resistance. Nature, 2019, 574 (7779), 549–552.

https://doi.org/10.1038/s41586-019-1662-9.

[77] Andersson, A. F.; Banfield, J. F. Virus Population Dynamics and Acquired Virus Resistance in

Natural Microbial Communities. Science (80-. )., 2008, 320 (5879), 1047–1050.

https://doi.org/10.1126/science.1157358.



https://doi.org/10.1128/IAI.00926-19.


203

[79] Baba, T.; Ara, T.; Hasegawa, M.; Takai, Y.; Okumura, Y.; Baba, M.; Datsenko, K. A.; Tomita, M.;

Wanner, B. L.; Mori, H. Construction of Escherichia Coli K-12 in-Frame, Single-Gene Knockout

Mutants: The Keio Collection. Mol. Syst. Biol., 2006, 2, 2006.0008.

https://doi.org/10.1038/msb4100050.

[80] Qimron, U.; Marintcheva, B.; Tabor, S.; Richardson, C. C. Genomewide Screens for Escherichia

Coli Genes Affecting Growth of T7 Bacteriophage. Proc. Natl. Acad. Sci. U. S. A., 2006, 103 (50),

19039–19044. https://doi.org/10.1073/pnas.0609428103.

[81] Maynard, N. D.; Birch, E. W.; Sanghvi, J. C.; Chen, L.; Gutschow, M. V.; Covert, M. W. A Forward-

Genetic Screen and Dynamic Analysis of Lambda Phage Host-Dependencies Reveals an Extensive

Interaction Network and a New Anti-Viral Strategy. PLoS Genet., 2010, 6 (7), e1001017.


[82] Kortright, K. E.; Chan, B. K.; Turner, P. E. High-Throughput Discovery of Phage Receptors Using

Transposon Insertion Sequencing of Bacteria. Proc. Natl. Acad. Sci. U. S. A., 2020, 117 (31), 18670–

18679. https://doi.org/10.1073/pnas.2001888117.

[83] Borin, J. M.; Avrani, S.; Barrick, J. E.; Petrie, K. L.; Meyer, J. R. Coevolutionary Phage Training

Leads to Greater Bacterial Suppression and Delays the Evolution of Phage Resistance. Proc. Natl.

Acad. Sci. U. S. A., 2021, 118 (23), e2104592118. https://doi.org/10.1073/pnas.2104592118.



[85] Weigele, P.; Raleigh, E. A. Biosynthesis and Function of Modified Bases in Bacteria and Their

Viruses. Chem. Rev., 2016, 116 (20), 12655–12687. https://doi.org/10.1021/acs.chemrev.6b00114.

[86] Łobocka, M.; Dąbrowska, K.; Górski, A. Engineered Bacteriophage Therapeutics: Rationale,

Challenges and Future. BioDrugs, 2021, 35 (3), 255–280. https://doi.org/10.1007/s40259-021-

00480-z.

[87] Bretaudeau, L.; Tremblais, K.; Aubrit, F.; Meichenin, M.; Arnaud, I. Good Manufacturing Practice

(GMP) Compliance for Phage Therapy Medicinal Products. Front. Microbiol., 2020, 11, 1161.


[88] Loponte, R.; Pagnini, U.; Iovane, G.; Pisanelli, G. Phage Therapy in Veterinary Medicine.

Antibiotics, 2021, 10 (4), 421. https://doi.org/10.3390/antibiotics10040421.





[90] Kittler, S.; Mengden, R.; Korf, I. H. E.; Bierbrodt, A.; Wittmann, J.; Plötz, M.; Jung, A.; Lehnherr,

T.; Rohde, C.; Lehnherr, H.; et al. Impact of Bacteriophage-Supplemented Drinking Water on the E.

Coli Population in the Chicken Gut. Pathogens, 2020, 9 (4), 293.

https://doi.org/10.3390/pathogens9040293.

[91] Chibani-Chennoufi, S.; Sidoti, J.; Bruttin, A.; Kutter, E.; Sarker, S.; Brüssow, H. In Vitro and in

Vivo Bacteriolytic Activities of Escherichia Coli Phages: Implications for Phage Therapy.

Antimicrob. Agents Chemother., 2004, 48 (7), 2558–2569. https://doi.org/10.1128/AAC.48.7.2558-

2569.2004.

[92] Jessup, C. M.; Bohannan, B. J. M. The Shape of an Ecological Trade-off Varies with Environment.

Ecol. Lett., 2008, 11 (9), 947–959. https://doi.org/10.1111/j.1461-0248.2008.01205.x.

[93] Vale, P. F.; Lafforgue, G.; Gatchitch, F.; Gardan, R.; Moineau, S.; Gandon, S. Costs of CRISPR -





204



[95] Gurney, J.; Brown, S. P.; Kaltz, O.; Hochberg, M. E. Steering Phages to Combat Bacterial Pathogens.

Trends Microbiol., 2020, 28 (2), 85–94. https://doi.org/10.1016/j.tim.2019.10.007.

[96] Ricci, V.; Piddock, L. J. V. Exploiting the Role of TolC in Pathogenicity: Identificatio n of a

Bacteriophage for Eradication of Salmonella Serovars from Poultry. Appl. Environ. Microbiol.,

2010, 76 (5), 1704–1706. https://doi.org/10.1128/AEM.02681-09.

[97] Chan, B. K.; Sistrom, M.; Wertz, J. E.; Kortright, K. E.; Narayan, D.; Turner, P. E. Phage Selection

Restores Antibiotic Sensitivity in MDR Pseudomonas Aeruginosa. Sci. Rep., 2016, 6, 26717.

https://doi.org/10.1038/srep26717.

[98] León, M.; Bastías, R. Virulence Reduction in Bacteriophage Resistant Bacteria. Front. Microbiol.,

2015, 6 (343), 1–7. https://doi.org/10.3389/fmicb.2015.00343.

[99] Meaden, S.; Paszkiewicz, K.; Koskella, B. The Cost of Phage Resistance in a Plant Pathogenic

Bacterium Is Context-Dependent. Evolution (N. Y)., 2015, 69 (5), 1321–1328.

https://doi.org/10.1111/evo.12652.

[100] Westra, E. R.; Van houte, S.; Oyesiku-Blakemore, S.; Makin, B.; Broniewski, J. M.; Best, A.; Bondy-

Denomy, J.; Davidson, A.; Boots, M.; Buckling, A. Parasite Exposure Drives Selective Evolution of

Constitutive versus Inducible Defense. Curr. Biol., 2015, 25 (8), 1043–1049.

https://doi.org/10.1016/j.cub.2015.01.065.

[101] Tagliaferri, T. L.; Jansen, M.; Horz, H. P. Fighting Pathogenic Bacteria on Two Fronts: Phages and

Antibiotics as Combined Strategy. Front. Cell. Infect. Microbiol., 2019, 9, 22.

https://doi.org/10.3389/fcimb.2019.00022.

[102] He, S.; He, H.; Chen, Y.; Chen, Y.; Wang, W.; Yu, D. In Vitro and in Vivo Analysis of Antimicrobial

Agents Alone and in Combination against Multi-Drug Resistant Acinetobacter Baumannii. Front.


[103] Liu, C. G.; Green, S. I.; Min, L.; Clark, J. R.; Salazar, K. C.; Terwilliger, A. L.; Kaplan, H. B.;

Trautner, B. W.; Ramig, R. F.; Maresso, A. W. Phage-Antibiotic Synergy Is Driven by a Unique

Combination of Antibacterial Mechanism of Action and Stoichiometry. MBio, 2020, 11 (4), e01462-


[104] Torres-Barceló, C.; Hochberg, M. E. Evolutionary Rationale for Phages as Complements of

Antibiotics. Trends Microbiol., 2016, 24 (4), 249–256. https://doi.org/10.1016/j.tim.2015.12.011.

[105] Segall, A. M.; Roach, D. R.; Strathdee, S. A. Stronger Together? Perspectives on Phage-Antibiotic

Synergy in Clinical Applications of Phage Therapy. Curr. Opin. Microbiol., 2019, 51, 46–50.

https://doi.org/10.1016/j.mib.2019.03.005.




[107] Fernández, L.; Gutiérrez, D.; García, P.; Rodríguez, A. The Perfect Bacteriophage for Therapeutic

Applications—A Quick Guide. Antibiotics, 2019, 8 (3), 126.


[108] Nale, J. Y.; Clokie, M. R. Preclinical Data and Safety Assessment of Phage Therapy in Humans.

Curr. Opin. Biotechnol., 2021, 68, 310–317. https://doi.org/10.1016/j.copbio.2021.03.002.



[110] Benala, M.; Vaiyapuri, M.; Visnuvinayagam, S.; George, J. C.; Raveendran, K.; George, I.;

Mothadaka, M. P.; Badireddy, M. R. A Revisited Two-Step Microtiter Plate Assay: Optimization of

in Vitro Multiplicity of Infection (MOI) for Coliphage and Vibriophage. J. Virol. Methods, 2021,


205

294, 114177. https://doi.org/10.1016/j.jviromet.2021.114177.

[111] Pinto, A. M.; Faustino, A.; Pastrana, L. M.; Bañobre-López, M.; Sillankorva, S. Pseudomonas

Aeruginosa PAO 1 In Vitro Time–Kill Kinetics Using Single Phages and Phage Formulations—

Modulating Death, Adaptation, and Resistance. Antibiotics, 2021, 10 (7), 877.


[112] Bryson, K. J.; Garrido, D.; Esposito, M.; McLachlan, G.; Digard, P.; Schouler, C.; Guabiraba, R.;

Trapp, S.; Vervelde, L. Precision Cut Lung Slices: A Novel Versatile Tool to Examine Host -

Pathogen Interaction in the Chicken Lung. Vet. Res., 2020, 51, 2. https://doi.org/10.1186/s13567-

019-0733-0.

[113] Jończyk-Matysiak, E.; Łodej, N.; Kula, D.; Owczarek, B.; Orwat, F.; Międzybrodzki, R.; Neuberg,

J.; Bagińska, N.; Weber-Dąbrowska, B.; Górski, A. Factors Determining Phage Stability/Activity:

Challenges in Practical Phage Application. Expert Rev. Anti. Infect. Ther., 2019, 17 (8), 583–606.

https://doi.org/10.1080/14787210.2019.1646126.

[114] García, R.; Latz, S.; Romero, J.; Higuera, G.; García, K.; Bastías, R. Bacteriophage Production

Models: An Overview. Front. Microbiol., 2019, 10, 1187.


[115] Mosimann, S.; Desiree, K.; Ebner, P. Efficacy of Phage Therapy in Poultry: A Systematic Review

and Meta-Analysis. Poult. Sci., 2021, 100 (12), 101472. https://doi.org/10.1016/j.psj.2021.101472.

Summary

207

Summary

Excess use of antimicrobials and release into the environment for over half a century have

generated a constant selective pressure for resistant bacterial strains. Consequently, we are

facing a worldwide antibiotic resistance challenge with increasing numbers of bacterial

infection becoming difficult to treat once again. Avian pathogenic Escherichia coli (APEC) is

one of the leading pathogens affecting poultry worldwide, and various multi-drug resistant

strains have been isolated. APEC strains with O-serogroup O1, O2, or O78 have been shown

to cause the majority of infections. The advances in the therapeutic use of the bacterial viruses,

bacteriophages (phages), have highlighted their potential use as an alternative or supplement

treatment against bacterial pathogens. However, we are only just beginning to understand the

diversity of phages, and the use of phages in therapy requires a detailed characterisation of the

candidate phages prior to their application to ensure that they have the expected potential to

kill pathogenic bacteria and have therapeutic effects, while minimising negative environmental

modifications. The understanding of the phage-host interactions has shown to be essential for

the development and application of a successful phage therapy. Th is PhD dissertation

contributes to our understanding of E. coli-infecting phage (coliphage)-host interaction for

phage therapy against APEC by determine in vitro growth dynamics as well as underlying

mechanism of phage resistance.

Summary

208

Phages are the most abundant organisms on Earth and can be found in every explored

ecosystem. In recent years, advances in sequencing and bioinformatics have broadened our

understanding of phage diversity, taxonomy, host-specificity, population structure and

genomic evolution. In chapter 3.1, we characterised coliphages isolated from poultry faeces.

The characterisation included phenotypic characterisation of the phage morphology and

genetic analysis of the phage genome. All isolated coliphages belonged to the Caudovirales

order (tailed phages), which comprise ~96% of the phages isolated to date. Phylogenetic

analysis based on conserved “marker” genes as well as full draft genomes was performed and

grouped phages according to genome size, G+C content and, phage subfamily. While the

phages were isolated from an otherwise similar ecosystem, a great diversity was observed

amongst them highlighting the mosaic nature of the phage genomes as well as the continuous

evolution. In accordance with previous studies, most of the phage genes products were

annotated as hypothetical proteins of unknown function. However, for all phages no genes

encoding known virulence- or antibiotic resistance- associated genes as well as other undesired

characteristics (for phage therapy) were detected.

Phage therapy represents a supportive to antibiotics to control bacterial pathogens. However,

current research indicates that there are still shortcomings in our understanding of phage

therapy as the in vitro results do not always correlate with the in vivo results. It has become

clear that knowledge on the phage-host interaction is essential to select and/or construct phages

with the desired host spectrum and activity, and hereby a more reliable in vitro and in vivo

outcome. Therefore, in chapter 3.2, we established an in vitro model to determine the

coliphage-host interaction and population growth dynamics. The coliphages (characterised in

chapter 3.1) were co-cultured with each of 10 APEC strains with O-serogroup O1, O2, or O78.

Growth dynamics were classified based on optical density (OD) of bacterial growth using a

combined exploratory and statistical approach. Growth dynamic patterns were defined as

resistant, susceptible, or in-between. Various factors affecting the phage pharmacokinetics

(PK) and pharmacodynamics (PD) have been described using mathematical and experimental

models. In this study, the influence of the three parameters: phage type, APEC strain, and

multiplicity of infection (MOI) were determined, identifying the MOI as the factor with the

least significant effect. In accordance with previous finding, different phage types showed

different host range, Tevenvirinae exhibiting the broadest and Ounavirinae not able to infect

any of the APEC strains. The established in vitro model was not only used to gain a better

Summary

209

understanding of the phage PK/PD but also provided a fast quantitative screening method for

candidate phages against a target bacterial pathogen.

One of the major concerns of phage therapy is the emergence of phage-resistant bacterial

mutants. Bacteria can develop resistance against phage through various mechanisms, including

modification of phage receptor-encoding genes and innate immune systems (such as CRISPR-

Cas), each associated with a different level of fitness cost for the bacterial strain. In chapter

3.3 we determined the factors involved in APEC in vitro phage resistance. Spontaneous phage

resistant strains were obtained from liquid co-cultures of the susceptible combination of

virulent Tevenvirinae coliphages and O1 APEC strain (determined in chapter 3.2). Whole-

genome sequence (WGS) analysis revealed that one or more single nucleotide polymorphisms

(SNPs) were detected in the bacterial core genome and/or that one or more genes were lost

(partial or complete). Genes affected by these genetic changes included genes known to be

involved in phage resistance through adsorption inhibition, including outer membrane protein

A (OmpA), lipopolysaccharide (LPS)-, O-antigen-, or cell wall-related genes as well as genes

not previously linked to phage resistance, including several hypothetical genes. Using bacterial

growth as an indicator, we determined the fitness cost associated with the genetic change(s)

detected. For several phage resistant mutants decrease (up to 65%) in overall growth was

detected. However, the magnitude of such fitness cost may vary in more complex in vivo

environment. Interestingly, genetic changes in genes encoding hypothetical proteins with

unknown function were one of the most prevalent and associated with a great fitness cost. As

such, these genes could serve as targets in future studies and potentially be exploited in phage

therapy.

In conclusion, this thesis provided novel insights into the coliphage diversity in the intestine

of poultry as well as the overall coliphage diversity. Moreover, a well-characterised collection

of coliphages was established, some of which with desired properties for phage therapy. By

studying the phage-host (APEC) interaction, much needed knowledge essential for a better

understanding of growth dynamics and the underlying mechanisms of phage resistance was

obtained. Although the full complexity of the interactions cannot be captured in vitro, this

knowledge is essential for the development of a more reliable, and hereby, a future successful

phage therapy against APEC.

Samenvatting

211

Samenvatting

Overmatig gebruik van antimicrobiële middelen en het vrijkomen van deze producten in de

omgeving gedurende meer dan een halve eeuw, hebben geleid tot een constante selectiedruk

voor resistente bacteriën. Bijgevolg worden we wereldwijd steeds vaker geconfronteerd met

de uitdagingen die met antibioticaresistentie gepaard gaan en waarbij een toenemend aantal

bacteriële infecties moeilijk te behandelen wordt. Aviaire pathogene Escherichia coli (APEC)

is wereldwijd een van de belangrijkste bacteriële pathogenen in pluimvee en er zijn reeds

verschillende multiresistente stammen geïsoleerd. De meerderheid van deze infecties zou

veroorzaakt worden door APEC-stammen met O-serogroep O1, O2 of O78. De vooruitgang in

het therapeutisch gebruik van bacteriële virussen of bacteriofagen (fagen), hebben duidelijk

potentieel als alternatieve of aanvullende behandeling tegen bacteriële pathogenen. We

beginnen de diversiteit van deze fagen echter nog maar net ten volle te begrijpen. Voorafgaand

aan hun toepassing, vereist het gebruik van fagen in therapie een gedetailleerde karakterisering

van de kandidaat-fagen. Dit is nodig om ervoor te zorgen dat ze hun potentieel om pathogene

bacteriën te doden en therapeutisch gunstige effecten te veroorzaken, ten volle kunnen benutten

en negatieve effecten op het milieu worden beperkt. Het verstaan van de faag-gastheer

interacties is essentieel gebleken voor de ontwikkeling en toepassing van een succesvolle

Samenvatting

212

faagtherapie. Deze doctoraatsstudie vergroot ons inzicht in de interactie tussen de E. coli-

infecterende faag (colifaag) en de gastheer voor faagtherapie tegen APEC, met behulp van in

vitro groeidynamieken, alsook door het onderliggende mechanisme van faagresistentie te

bepalen.

Fagen zijn de meest voorkomende organismen op Aarde en zijn te vinden in elk ecosysteem.

In de afgelopen jaren hebben vooruitgangen in sequenering en bio-informatica ervoor gezorgd

dat we faagdiversiteit, taxonomie, gastheerspecificiteit, populatiestructuur en genomische

evoluties nog beter verstaan. In hoofdstuk 3.1 werden colifagen gekarakteriseerd die werden

geïsoleerd uit uitwerpselen van pluimvee. De karakterisering omvatte fenotypische

karakterisering van de faagmorfologie en genetische analyse van het faaggenoom. Alle

geïsoleerde colifagen behoorden tot de orde van de Caudovirales (staartfagen), die ~96% van

de tot nu toe geïsoleerde fagen omvat. Met behulp van fylogenetische analyse op basis van

geconserveerde merker-genen en volledige conceptgenomen werden fagen gegroepeerd

volgens genoomgrootte, G+C-gehalte en faagsubfamilie. Hoewel de fagen werden geïsoleerd

uit een gelijkaardig ecosysteem, werd er een grote diversiteit onder hen waargenomen, wat

wijst op de mozaïek-structuur van de faaggenomen en een continue evolutie. In

overeenstemming met eerdere studies werden de meeste faag genenproducten geannoteerd als

hypothetische eiwitten met onbekende functie. Bij geen enkele van de bestudeerde fagen

werden echter genen gedetecteerd die coderen voor bekende virulentie- of

antibioticaresistentie-geassocieerde genen, evenals andere ongewenste kenmerken (voor

faagtherapie).

Faagtherapie ondersteunt antibiotica om bacteriële pathogenen onder controle te houden.

Huidig onderzoek geeft echter aan dat we faagtherapie nog altijd niet ten volle begrijpen,

aangezien de in vitro resultaten niet altijd correleren met de in vivo resultaten. Kennis over de

faag-gastheer interactie is essentieel om fagen met het gewenste gastheerspectrum en -activiteit

te selecteren en/of te construeren, en hiermee een betrouwbaarder in vitro en in vivo resultaat

te genereren. Daarom werd er in hoofdstuk 3.2 een in vitro model opgesteld om de interactie

tussen colifaag en gastheer en de populatiegroei dynamiek te bepalen. De colifagen (besproken

in hoofdstuk 3.1) werden samen opgekweekt met elk van de 10 APEC-stammen met O-

serogroep O1, O2 of O78. De groeidynamiek werd geclassificeerd op basis van optische

dichtheid (OD) van bacteriegroei met behulp van een gecombineerde verkennende en

statistische benadering. Dynamische groeipatronen werden gedefinieerd als resistent, vatbaar

of daartussenin. Verschillende factoren die de farmacokinetiek van faag (PK) en

Samenvatting

213

farmacodynamiek (PD) beïnvloeden, werden beschreven met behulp van wiskundige en

experimentele modellen. In deze studie werd de invloed van de drie parameters: faagtype,

APEC-stam en multipliciteit van infectie (MOI) bepaald, waarbij de MOI werd geïdentificeerd

als de factor met slechts weinig of geen effect. In overeenstemming met eerdere bevindingen

vertoonden verschillende faagtypen verschillende mogelijke gastheren. Tevenvirinae

vertoonde het grootste gastheerbereik en Ounavirinae waren niet in staat om één van de APEC-

stammen te infecteren. Het gebuikte in vitro model werd niet alleen gebruikt om een beter

begrip te verkrijgen van de faag PK/PD, maar bood ook een snelle kwantitatieve

screeningsmethode voor kandidaat-fagen tegen een bacteriële doelwitpathogeen.

Opkomst van faagresistente bacteriële mutanten is momenteel één van de grootste zorgen.

Bacteriën kunnen resistentie tegen fagen ontwikkelen via verschillende mechanismen,

waaronder modificatie van genen die faagreceptoren coderen en via hun aangeboren

immuunsysteem (zoals CRISPR-Cas), elke aanpassing resulteert in een eigen fitnesskost voor

de bacteriestam. In hoofdstuk 3.3 werden de factoren bepaald die betrokken zijn bij de in vitro

faagresistentie van APEC. Spontane faagresistente stammen werden verkregen uit vloeibare

co-culturen van de gevoelige combinatie van virulente Tevenvirinae-colifagen en O1 APEC-

stam (bepaald in hoofdstuk 3.2). Via whole-genome sequence (WGS)-analyse werden een of

meer single nucleotide polymorphisms (SNP's) gedetecteerd in het bacteriële kerngenoom

en/of werd ontdekt dat een of meerdere genen verloren gingen (gedeeltelijk of volledig). Genen

die bij deze veranderingen betrokken zijn, omvatten genen waarvan bekend is dat ze betrokken

zijn bij faagresistentie door middel van adsorptieremming, waaronder buitenmembraaneiwit A

(OmpA), lipopolysaccharide (LPS)-, O-antigeen- of celwandgerelateerde genen, evenals genen

die voorheen niet gelinkt werden aan faagresistentie, waaronder verschillende hypothetische

genen. Met behulp van bacteriële groei als indicator werd de “fitnesscost” bepaald die gelinkt

is aan deze gedetecteerde genetische verandering(en). Voor verschillende faagresistente

mutanten werd een verminderde (tot 65%) totale groei gedetecteerd. De omvang van dergelijke

“fitnesscost” kan echter verschillen in een complexere in vivo-omgeving. Interessant is dat

veranderingen in genen die coderen voor hypothetische eiwitten met onbekende functie, een

van de meest voorkomende waren en dat deze ook geassocieerd werden met een hoge

“fitnesscost”. Deze genen kunnen daarom verder onderzocht worden in toekomstige studies en

mogelijks ook benut worden bij faagtherapie.

Ter conclusie kan er gesteld worden dat er nieuwe inzichten zijn verschaft in de

colifagendiversiteit in de darmen van pluimvee en de algehele colifagendiversiteit. Bovendien

Samenvatting

214

werd een goed gekarakteriseerde collectie colifagen bekomen, waarvan sommige met gewenste

eigenschappen voor faagtherapie. Door de faag-gastheer (APEC) interactie te bestuderen, werd

de nodige kennis verkregen die essentieel is voor een beter begrip van de groeidynamiek en de

onderliggende mechanismen van faagresistentie. Hoewel de volledige complexiteit van de

interacties niet in vitro kan worden vastgelegd, is deze kennis essentieel voor de ontwikkeling

van een meer betrouwbare en daarmee een toekomstige succesvolle faagtherapie tegen APEC.

Curriculum Vitae

215

Curriculum Vitae

Patricia Espenhain Sørensen was born on July 4, 1992 in Gentofte, Denmark. In 2018, she

obtained a Master of Science degree in Biology-Biotechnology from University of

Copenhagen, Denmark. From June 2018, she joined the Department of Pathology,

Bacteriology and Poultry Diseases at the Faculty of Veterinary Medicine as a doctoral student,

where she spent three years researching the interactions of bacteriophages and avian pathogenic

Escherichia coli (APEC). She is the first author of three publications in international journals

and has presented her research results at international conferences.

The research project is a part of the EU’s ambitious Horizon 2020 Marie Skłodowska -Curie

Initial Training Network: Combatting Antimicrobial Resistance Training Network

(CARTNET), and is conducted in a collaboration between Ghent University, University of

Copenhagen, Denmark, Ross University School of Veterinary Medicine (RUSVM), St. Kitts,

and Statens Serum Institut, Denmark.

Bibliography

217

Bibliography

Sørensen, P.E., Van Den Broeck, W., Kiil, K., Jasinskyte, J., Moodley, A., Garmyn, A.,

Ingmer, H., and P. Butaye. 2020. New insights into the biodiversity of coliphages in the

intestine of poultry. Sci Rep 10, 15220. DOI: 10.1038/s41598-020-72177-2.

Sørensen, P. E., Ng, D., Duchateau, L., Ingmer, H., Garmyn, A., and P. Butaye. 2021.

Classification of in vitro phage–host population growth dynamics. Microorganisms 9, 2470.

DOI: 10.3390/microorganisms9122470

Sørensen, P. E., Baig, S., Stegger, M., Ingmer, H., Garmyn, A., and P. Butaye. 2021.

Spontaneous phage resistance in avian pathogenic Escherichia coli. Front. Microbiol. 12,

782757. DOI: 10.3389/fmicb.2021.782757

Conference contributions

219

Conference contributions

ASM Microbe 2020 Online

Sørensen, P.E., Garmyn, A., Ingmer, H., and P. Butaye. Dynamics of Bacteriophage-Host

Interactions. ePoster presentation. July 20, 2020.

Sørensen, P.E., Garmyn, A., Kiil, K., Jasinskyte, D., Moodley, A., Ingmer, H., and P. Butaye.

Novel Insights into the Biodiversity of Coliphages in the Intestine of Poultry. ePoster

presentation. July 20, 2020.

World Microbe Forum 2021 (online)

Sørensen, P.E., Baig, S., Stegger, M., Garmyn, A., Ingmer, H., and P. Butaye. Phage-Host

Interactions: In Vitro Generated E. coli Phage Resistance . iPoster presentation. June 20, 2021.

Acknowledgements

221

Acknowledgements

The past three and a half years have been quite a journey, which ultimately resulted in this

dissertation. All of this would not have been possible without the help and support of many

different people. I would like to thank everyone that contributed, directly or indirectly, to me

achieving my doctoral degree.

First and foremost, I would like to thank my three supervisors. Prof. dr. Patrick Butaye, thank

you for giving me the opportunity to do this PhD, and for all your guidance, patience, and

encouragement. Prof. dr. An Garmyn. Thank you for always making time for my questions and

for reading my reports and manuscripts quickly and thoroughly and for the valuable feedback.

Prof. dr. Hanne Ingmer, despite the physical distance during my PhD, you have always showed

interest in my work. Thank you for this, and for giving me support when I needed it.

I would also like to thank the members of my examination committee, chair Prof. dr. Niek

Sanders, Prof. dr. Gunther Antonissen, Dr. Ilias Chantziaras, dr. Steven Van Borm, Prof. dr.

Rob Lavigne, and Prof. dr. Felix Toka. Thank you for taking your time to read this thesis, for

Acknowledgements

222

your genuine interest, constructive questions, and valuable suggestions to improve this

dissertation.

A special thanks to the co-authors of my publications, Dziuginta, Arshnee, Luc, Marc, Sharmin,

Duncan, and Kristoffer. Thank you for all your assistance and good collaborations.

I want to thank the CARTNET consortium for the great training and network meetings and for

staying connected even when a worldwide pandemic made it challenging. Especially Duncan.

Thanks for all the crash courses in bioinformatics and for helping me even as you yourself were

busy and out of time in your own PhD. To my WP3 fellows, Frida, Helena, Anaëlle, and Emilia.

Our professional collaboration might have been limited, but I am truly grateful for all the nice

moments and conversations we have shared.

My gratitude also goes to current and past members from the Department of Pathobiology,

Pharmacology and Zoological Medicine for providing such a pleasant environment in and

outside of the lab. I am grateful to Serge for his cooperation during the COVID-19 lock-down

and for assisting my lab work when restrictions kept me from going. To Marleen, thank you

for always being so accommodating and helpful whenever needed. I acknowledge Liesbeth for

performing the electron microscopy (TEM) reported in chapter 3.1, and Kristof, for assisting

with a translation when my Dutch skills were insufficient. To my office members, Jasmien,

Evelien, Yani, and Katrien, thank you for the nice chats in-between lab experiments. Evy, thank

you for always making time, even though your schedule is always full. To Jill, Tessa, Ilse,

Martina, Alessandra, Lore, Silvio, and Svieta, you have all been a special part of my journey

in Belgium and I am forever grateful for all the great moments we have shared.

Thank you to the members of the RUSVM phage group. Especially Andreas for his guidance

and in the beginning of the PhD and Jake for assisting with the phage work on St. Kitts.

Furthermore, I would like to thank my St. Kitts Island family for making my stay in St. Kitts

truly special. Thank you for the many hours of Monday beach volley, snorkel tours, bonfires

and the great conversations and moments we spent together.

I want to thank my family and my closest friends for their unconditional love and support in

whatever I do. Thank you for being there when needed and for taking your time to come visit.

Finally, I would like to thank my partner Cecilie. Thank you for your love, patience, and

support during this rollercoaster ride and for always believing in me. Thank you for following

me across the world (twice) and being my person through it all.

host interaction for phage therapy against avian pathogenic

Documents